EXISTING USES AND THE LIMITS OF
LAND USE REGULATIONS
CHRISTOPHER SERKIN*
This Article identifies property law’s special protection for existing uses, explores
possible justifications for this protection, and argues that none can support the
strong protection that existing uses currently enjoy. Various land use doctrines—
from zoning to the vested rights doctrine to amortization rules for prior nonconforming uses—assume that the government cannot eliminate existing uses without
paying compensation. The Article asks whether this result is compelled either by
constitutional rules or by normative considerations. Neither the Takings Clause nor
the Due Process Clause requires this level of protection for existing uses. Normatively, many obvious-seeming justifications dissolve on closer inspection. Objections grounded in underlying principles of fairness and reliance are not
conceptually different for regulations prohibiting future uses than for regulations of
existing uses. Nor is the extent of economic loss necessarily greater for one than for
the other, even though regulations of existing uses create out-of-pocket costs
whereas regulations of future uses only implicate forgone profits. In fact, none of
the possible explanations for the special treatment of existing uses actually justifies
their protection. This Article ultimately concludes that existing uses should not be
entitled to any special judicial protection but instead should be subject to the same
takings and due process analyses that apply to all regulation and governmental
action.
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I. EXISTING USES IN THE LAW . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Retroactive Zoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Prior Nonconforming Uses and Amortization . . . . . . .
C. The Vested Rights Doctrine . . . . . . . . . . . . . . . . . . . . . . . . .
D. The Nuisance Exception . . . . . . . . . . . . . . . . . . . . . . . . . . . .
II. CONSTITUTIONAL PROTECTION FOR EXISTING USES . . . .
A. Constitutional Underpinnings . . . . . . . . . . . . . . . . . . . . . . .
1223
1232
1232
1235
1238
1240
1242
1243
* Copyright © 2009 by Christopher Serkin, Associate Professor, Brooklyn Law School;
Visiting Associate Professor, the University of Chicago Law School. Thanks to participants at the 2008 Stanford/Yale Junior Faculty Forum, the 2008 Property Theory Workshop
at New York University, and the 2008 Property Works in Progress Conference at the
University of Colorado, and in particular to Tino Cuéllar, Bill Nelson, Amnon Lehavi, and
Rachel Godsil. Special thanks go to Vicki Been, Michael Cahill, John Echeverria, Robert
Ellickson, Jill Fisch, William Fischel, Daryl Levinson, Richard Revesz, Nelson Tebbe, and
Katrina Wyman for their extremely helpful comments on earlier drafts. Thanks also to
participants in faculty workshops at the University of Chicago Law School, the University
of Michigan Law School, New York University School of Law, the University of
Pennsylvania Law School, Vanderbilt University Law School, and Brooklyn Law School.
The piece benefited from early conversation with David Post and David Reiss. Jim Krier
deserves special mention for his significant input and encouragement along the way. The
Brooklyn Law School Dean’s Summer Research Stipend provided financial support for
this project. David Schnakenberg provided outstanding research assistance.
1222
November 2009]
EXISTING USES
B. The Takings Clause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. The Due Process Clause . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. The Relationship Between Takings and Due
Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. NORMATIVE ACCOUNTS OF EXISTING USE
PROTECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Antiretroactivity and Existing Uses . . . . . . . . . . . . . . . . . .
B. Nature and Extent of the Loss . . . . . . . . . . . . . . . . . . . . . .
1. Fairness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Extent of Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Waste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Personhood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Protecting the Status Quo . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Objective Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Evidence of Subjective Expectations . . . . . . . . . . . . .
D. Political Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IV. LIMITING EXISTING USE PROTECTION . . . . . . . . . . . . . . . . . .
A. The High Costs of Existing Use Protection . . . . . . . . . .
B. The Illusory Benefits of Existing Use Protection . . . .
C. Designing Appropriate Protection for Existing Uses .
CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1223
1249
1256
1258
1261
1262
1265
1266
1267
1270
1273
1274
1274
1275
1277
1281
1283
1287
1288
1290
INTRODUCTION
Existing uses occupy a special place in property law. A use, once
established, is imbued with an expectation that it may continue to
exist, even in the face of regulatory change. Once built, a building
becomes all but immune from subsequently enacted zoning rules.
Indeed, courts often reason that existing uses are constitutionally protected from government regulations.1 A number of land use doc1 See, e.g., Hooper v. City of St. Paul, 353 N.W.2d 138, 140 (Minn. 1984) (referring to
“constitutional protection afforded existing uses” (citing County of Freeborn v. Claussen,
203 N.W.2d 323, 325 (Minn. 1972))); Rudolf Steiner Fellowship Found. v. De Luccia, 685
N.E.2d 192, 197 (N.Y. 1997) (Bellacosa, J., dissenting) (“It is the law of this state that
nonconforming uses or structures, in existence when a zoning ordinance is enacted, are, as
a general rule, constitutionally protected and will be permitted to continue, notwithstanding the contrary provisions of the ordinance.” (citing People v. Miller, 106 N.E.2d 34,
35 (N.Y. 1952))); Harbison v. City of Buffalo, 152 N.E.2d 42, 44 (N.Y. 1958) (“When
zoning ordinances are initially adopted to limit permissible uses of property, or when property is rezoned . . . to prevent uses of property previously allowed, a degree of protection is
constitutionally required to be given owners of property then using their premises in a
manner forbidden by the ordinance.”); Snake River Brewing Co. v. Town of Jackson, 39
P.3d 397, 403 (Wyo. 2002) (“When a zoning ordinance is enacted, it cannot outlaw previously existing non-conforming uses. This right to continue a non-conforming use is a
vested property right, protected by statute, and by both federal and state constitutions.”
(citations and internal quotation marks omitted)).
1224
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
trines, from vested rights to amortization of prior nonconforming uses,
are based fundamentally on an unarticulated but firm assumption that
the government cannot interfere with existing uses, subject only to
some specific exceptions.2 There is, in short, a strong background rule
running throughout the law of property that existing uses are entitled
to protection from the government. This Article argues that the law
actually overprotects existing uses and that neither constitutional doctrine nor normative considerations justify the protection courts currently provide.
The legal literature has largely assumed that existing uses are
entitled to protection and has almost wholly failed to examine the
basis for that assumption.3 At first glance, this is hardly surprising.
Most people share an intuition that existing uses should be protected,
and the law generally complies. Consider zoning first. A local government enacting a new zoning ordinance must almost always grandfather existing uses.4 The grocery store in a newly zoned residential
area is allowed to stay in business.5 New height, area, or use restrictions are imposed prospectively only. Try even to imagine what it
would mean for a local government to force preexisting houses to conform to new setback requirements. Nonconforming homes would
either have to be picked up and moved back or simply torn down. It
2
For discussion of these doctrines, see infra Part I.
Some specific doctrines related to existing uses have received serious study, however.
For example, the doctrines governing the treatment of prior nonconforming uses, which
are the subject of Part I.B, infra, have received significant attention.
4 See, e.g., MASS. GEN. LAWS ANN. ch. 40A, § 6 (West 2004) (forbidding zoning ordinances to apply to structures “lawfully in existence or lawfully begun” before first notice of
new law, but requiring their application to “any change or substantial extension of such
use”); N.J. STAT. ANN. § 40:55D-68 (West 2008) (“Any nonconforming use or structure
existing at the time of the passage of an ordinance may be continued upon the lot or in the
structure so occupied and any such structure may be restored or repaired in the event of
partial destruction thereof.”); Rourke v. Rothman, 859 N.E.2d 821, 822 n.5 (Mass. 2007)
(citing statute that “sets the floor for ‘grandfather’ protection in local zoning bylaws”); see
also Ferraro v. Bd. of Zoning Adjustment, 970 So. 2d 299, 310 (Ala. Civ. App. 2007)
(“Generally, nonconforming uses of property are ‘grandfathered’ under zoning ordinances
and not lost unless the owner abandons that use.”); Manhattan, Inc. v. Shelby County, No.
W2006-02017-COA-R3-CV, 2008 WL 639791, at *7 (Tenn. Ct. App. Mar. 11, 2008) (noting
that local zoning ordinances “specifically permit a lawful use in existence at the time of any
zoning change to continue after the change, i.e., ‘grandfathered’ nonconforming use, with
certain restrictions”).
5 See Ex parte Ala. Alcoholic Beverage Control Bd., 819 So. 2d 50, 52–53 (Ala. 2001)
(upholding Control Board’s denial of expansion of nonconforming use but recognizing
store as valid nonconforming use allowed to operate); Brewster v. Fayette County Bd. of
County Comm’rs, No. W2003-01842-COA-R3-CV, 2005 WL 873224, at *1 (Tenn. Ct. App.
Apr. 14, 2005) (holding that previously “grandfathered” grocery store on property could
not reopen without demonstrating continuous use of property). This rule is subject to
various techniques—like amortization—that local governments can, in fact, use to eliminate existing uses. See infra Part I.B (discussing amortization).
3
November 2009]
EXISTING USES
1225
is easy to understand the intuition that such a regulation must be
somehow unconstitutional.6
The calculus is not all so one-sided, however. The costs of protecting existing uses are extremely high. Working around existing uses
can severely limit the efficacy of zoning and other comprehensive land
use planning, potentially transforming prospective planning into a
mere description and codification of existing conditions.7 The original
justification for zoning was to separate incompatible uses of property;
the inapplicability of a new zoning ordinance to prior nonconforming
uses can preserve incompatible neighbors.8 Moreover, as explored in
an extensive literature on legal transitions, protecting people from
regulatory change may create perverse investment incentives.9 The
6 For a more emotional example, consider a co-op board seeking to enforce a new nopets policy. There is little doubt in most jurisdictions that such policies are permissible
prospectively, but it is another matter entirely to apply the policy against existing pets. See,
e.g., Winston Towers 200 Ass’n v. Saverio, 360 So. 2d 470, 470–71 (Fla. Dist. Ct. App. 1978)
(holding invalid attempt to fine condo owner for keeping puppy of dog grandfathered into
co-op’s no-pet policy), cited in ROBERT C. ELLICKSON & VICKI L. BEEN, LAND USE CONTROLS 597 (3d ed. 2005); Young v. Savinon, 492 A.2d 385, 390 (N.J. Super. Ct. App. Div.
1985) (invalidating no-pets provision in lease renewal as applied to existing dogs). For a
particularly funny examination of no-pet policies, see The Daily Show with Jon Stewart:
Rent Claws (Comedy Central television broadcast Mar. 29, 2004), available at http://www.
thedailyshow.com/video/index.jhtml?title=rent-claws&videoId=124322.
7 See 7 PATRICK J. ROHAN, ZONING AND LAND USE CONTROLS § 41.01[2], at 41–47
(2009) (“If the goal of [zoning] regulations was to ensure uniformity of all uses in a particular district, dissimilar existing uses would detract from that purpose as much as new
uses.”); Christopher Serkin, Local Property Law: Adjusting the Scale of Property Protection, 107 COLUM. L. REV. 883, 940 (2007) (“The extent of existing development in developed cities means that changes in applicable zoning must either include an enormous
number of nonconforming uses—potentially undermining the efficacy of the zoning
regime—or come with some plan for eliminating the nonconforming uses over time.”); see
also Cohen v. Duncan, No. Civ.A.2002-599, Civ.A.2001-380, 2004 WL 1351155, at *15 (R.I.
Super. Ct. June 9, 2004) (“The restrictive provisions of the [o]rdinance properly recognize
that ‘[n]onconforming uses are a thorn in the side of proper zoning and should not be
perpetuated any longer than necessary.’” (second alteration in original) (internal citation
omitted)).
8 The Supreme Court originally justified zoning as separating incompatible neighbors,
noting that a “nuisance may be merely a right thing in the wrong place, like a pig in the
parlor instead of the barnyard.” Vill. of Euclid v. Ambler Realty Co., 272 U.S. 365, 388
(1926); see also David P. Bryden, The Impact of Variances: A Study of Statewide Zoning,
61 MINN. L. REV. 769, 771–72 (1977) (describing use of variances as “especially dangerous
because they jeopardize zoning’s primary purpose: separating incompatible land uses”).
9 See infra note 40 and Part IV.A (discussing transitions literature). Academic writing
on legal transitions has tended to focus both on changes in tax rules and on changes in
property rules, with the latter squarely implicating the Takings Clause and the extent of
protection from regulatory change. See generally, e.g., DANIEL SHAVIRO, WHEN RULES
CHANGE: AN ECONOMIC AND POLITICAL ANALYSIS OF TRANSITION RELIEF AND RETROACTIVITY (2000) (discussing changes to tax rules); Louis Kaplow, An Economic Analysis of
Legal Transitions, 99 HARV. L. REV. 509, 517 (1986) (discussing changes to property rules).
1226
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
questions therefore arise whether zoning and other land use controls
could apply to existing uses and, if so, when.
The stakes here are very high. Existing use protection is bound
up intimately with the problem of grandfathering, a central issue in
environmental policy. Grandfathering existing uses can dramatically
limit the effectiveness of new environmental regulations.10 Protecting
existing uses can allow the dirtiest factories to keep polluting and the
worst offenders to avoid application of the new rules.11 The Clean Air
Act,12 which regulates stationary sources of air pollution, distinguishes
explicitly between preexisting and new sources of air pollution and
does not require the former to comply with its new emission standards.13 Interestingly, however, a stationary source that is subsequently modified within the meaning of the Act must then comply
with the new source performance standards—as if certain modifications transform it from a preexisting use into a prospective, future
use.14 The Clean Water Act,15 while taking a slightly different
approach, also applies differently to existing and new sources of
pollution.16
Other such examples are not confined to land use and environmental law. Many statutes incorporate protection for existing uses,
often by imposing different requirements for uses that predate enactment.17 For example, the Americans with Disabilities Act18 applies a
10 See generally, e.g., Jonathan Remy Nash & Richard L. Revesz, Grandfathering and
Environmental Regulation: The Law and Economics of New Source Review, 101 NW. U. L.
REV. 1677 (2007) (discussing harms of grandfather clauses in recent legislation).
11 See id. at 1729 (discussing how Bush administration policies exacerbated this
problem).
12 42 U.S.C. §§ 7401–7671q (2006).
13 42 U.S.C. § 7411(a)(3) (2006) (defining stationary source). Existing uses are subject
only to state implementation of ambient air quality requirements and not to the stricter
emissions reduction requirements for new sources, which are set by the federal government. See Deepa Varadarajan, Note, Billboards and Big Utilities: Borrowing Land-Use
Concepts To Regulate “Nonconforming” Sources Under the Clean Air Act, 112 YALE L.J.
2553, 2558 (2003).
14 42 U.S.C. § 7411(a)(2), (4) (2006) (defining new source as any source “the construction or modification of which is commenced after publication of regulations” and defining
modification as “any physical change in, or change in method of operation of, a stationary
source which increases the amount of air pollutant emitted”). This is like the converse of
the vested rights doctrine, discussed in Part I.C, infra.
15 33 U.S.C. §§ 1251–1387 (2006).
16 33 U.S.C. § 1316(d); see also Nash & Revesz, supra note 10, at 1728–29 (discussing
grandfathering provisions in Clean Water Act).
17 See Heidi Gorovitz Robertson, If Your Grandfather Could Pollute, So Can You:
Environmental “Grandfather Clauses” and Their Role in Environmental Inequity, 45 CATH.
U. L. REV. 131, 132–33 & nn.8–15 (1995) (providing examples of grandfather clauses from
various legal contexts); Steven Shavell, On Optimal Change, Past Behavior, and
Grandfathering, 37 J. LEGAL STUD. 37, 69–70 (2008) (same).
18 42 U.S.C. § 12101–12213 (2006).
November 2009]
EXISTING USES
1227
less restrictive accessibility standard to existing buildings than to new
ones.19 Similarly, the Wilderness Act20 grandfathers in a substantial
number of existing uses—grazing rights in particular.21 In 1995,
Florida enacted a new statute in response to a perceived failure of the
courts to afford sufficient protection to private property under the
state and federal takings clauses.22 That statute explicitly offers additional protection to existing uses, providing a new cause of action
when a government “has inordinately burdened an existing use of real
property or a vested right to a specific use of real property.”23 Other,
more targeted, examples are also easy to find.24
As a descriptive matter, much existing use protection can be
explained in purely political terms. It is a necessary concession to an
entrenched interest group—owners of existing uses—and is required
to secure a law’s passage.25 In fact, however, the politics do not
always line up so neatly in favor of protecting existing uses, especially
19 See 28 C.F.R. § 36.304 (2008) (implementing Title III of Americans with Disabilities
Act; requiring existing buildings to “remove architectural barriers in existing facilities” but
only “where such removal is readily achievable, i.e., easily accomplishable and able to be
carried out without much difficulty or expense”); see also John Grady & Damon Andrew,
Legal Implications of the Americans with Disabilities Act on Recreation Services: Changing
Guidelines, Structures, and Attitudes in Accommodating Guests with Disabilities, 13 J.
LEGAL ASPECTS SPORT 231, 235 (2003) (describing application of ADA to new and
existing facilities).
20 16 U.S.C. §§ 1131–1136 (2006).
21 See Mitchel P. McClaran, Livestock in Wilderness: A Review and Forecast, 20
ENVTL. L. 857, 858 (1990) (“A short-lived proposal to eliminate existing grazing, made
during the initial legislative process, was only an anomaly in an otherwise continuous history of grandfathering most grazing management structures and practices as acceptable
nonconforming uses in wilderness.”).
22 See Bert J. Harris, Jr., Private Property Rights Protection Act, FLA. STAT. ANN.
§ 70.001 (West 2006); see also Nicole S. Sayfie & Ronald L. Weaver, 1999 Update on the
Bert J. Harris Private Property Rights Protection, 73 FLA. B.J. 49 (1999) (discussing application of and alternatives to Florida statute).
23 FLA. STAT. ANN. § 70.001(2) (emphasis added). The statute defines an existing use,
in relevant part, as “an actual, present use or activity on the real property . . . or such
reasonably foreseeable, nonspeculative land uses which are suitable for the subject real
property and compatible with adjacent land uses.” Id. § 70.001(3)(b).
24 For example, Montana voters passed by referendum a law eliminating the use of
open pit mining with a cyanide reagent, but it grandfathered existing mines that already
had a permit. See MONT. CODE ANN. § 82-4-390 (2007). In 1980, Arizona adopted a new
law governing the use of groundwater that explicitly grandfathered existing uses of groundwater. See ARIZ. REV. STAT. ANN. § 45-462 (2003) (“The right to withdraw or receive and
use groundwater pursuant to this article is a grandfathered right.”).
25 See Robertson, supra note 17, at 132 (discussing grandfather clauses and observing
that “when a tough new law is proposed, affected industry lobbyists fight, often successfully, to exempt the existing industry from the new, presumably more stringent requirements”). But cf. Elizabeth Bogley Roth, Environmental Considerations in Hydroelectric
Licensing: California v. FERC (Dynamo Pond), 23 ENVTL. L. 1165, 1183 (1993) (“One
possible, but unarticulated, reason for the grandfather clause might be that Congress felt a
hydroelectric license represents some sort of property right.”).
1228
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
at the local level.26 There, and particularly in the context of land use
controls, local governments do sometimes seek to regulate away
existing uses. In response, courts have developed a variety of land use
doctrines to protect existing uses, many of which purport to be based
on constitutional limitations.27 This Article uses the specific context
of these local land use decisions to examine the underlying constitutional and normative justifications for the judicial protection of
existing uses.
Part I examines how current land use law strongly defends
existing uses, although this protection often lies below the surface and
is not immediately apparent. For example, amortization is an exception to the protection of existing uses that nevertheless clearly demonstrates the existence of the underlying rule.28 The doctrine of
amortization allows a local government to eliminate an existing use
without paying compensation so long as the use is allowed to remain
in place for some amount of time.29 But no amortization rule allows
the elimination of an existing use immediately, and, indeed, the duration of the amortization period is a matter of constitutional concern.30
The doctrine therefore implicitly assumes that existing uses cannot be
eliminated outright. Focusing only on the constitutionality of the
duration of an amortization provision is like debating the jail time
before deciding whether conduct was criminal. Other doctrines are
similarly influenced by excessive deference to existing uses.31 In fact,
the protection of existing uses runs like an underground current
throughout the law of property. Its likely source is either doctrinal or
normative, but both turn out to be unsatisfying.
Part II explores the doctrinal claim: that property law protects
existing uses because the Constitution requires it. This is at least
26
This point is considered in detail in Part III.D, infra.
Those doctrines are the subject of Part I, infra.
28 See Vill. of Valatie v. Smith, 632 N.E.2d 1264, 1266–67 (N.Y. 1994) (“Though the
amortization period is typically discussed in terms of protecting the owners’ financial interests, it serves more generally to protect ‘an individual’s interest in maintaining the present
use’ of the property.” (quoting Modjeska Sign Studios, Inc. v. Berle, 373 N.E.2d 255, 261
(N.Y. 1977))).
29 For a discussion of amortization, see infra Part I.B.
30 See, e.g., Mayor and Council of New Castle v. Rollins Outdoor Adver., Inc., 475 A.2d
355, 359 (Del. 1984) (holding that “reasonable, required amortization of such nonconforming uses as sign boards may be accomplished with due process of law, and that such
form of regulation does not necessarily constitute a compensable taking”); Friends of East
Fork v. Clark County, No. 33422-4-II, 2006 WL 1745032, at *1 n.1 (Wash. Ct. App. June 27,
2006) (“Lawful nonconforming uses may continue but, subject to constitutional limits, the
government may regulate or even terminate the use after a period of nonuse or a reasonable amortization period that allows the owner to recoup on investment.”).
31 These include, notably, the vested rights doctrine and the nuisance exception, both of
which are discussed at length in Part I.C–D, infra.
27
November 2009]
EXISTING USES
1229
partly true. However, there is hardly consensus about the constitutional source of the protection. In reviewing regulations of existing
uses, many courts fail to specify the precise constitutional basis for
their holdings.32 Often, courts simply assume that an existing use is
protected from regulation without further explanation.33
Today, the Takings Clause is the most likely source of protection,
but it cannot fully account for courts’ treatment of existing uses.
According to the Supreme Court in Penn Central, the Takings Clause
compels compensation when the government interferes with a property owner’s investment-backed expectations.34 The principal expectation to protect, according to the Supreme Court, is the existing use
of the property.35 In the years since Penn Central, however, courts
generally have not used this test to protect existing uses; instead, they
typically have used it to inquire whether a property owner’s expectations about a future use were reasonable.36 The other relevant basis
for liability under Penn Central is the diminution in value test, which
limits the amount a regulation can reduce property values before
requiring compensation.37 So long as some alternative use of the
32 See, e.g., Sterngass v. Woodbury, 433 F. Supp. 2d 351, 355 (S.D.N.Y. 2006) (“His only
right is to continue a pre-existing non-conforming use, without expansion or change.”);
Dublin v. Finkes, 615 N.E.2d 690, 692 (Ohio Ct. App. 1992) (finding that federal and state
constitutions both “recognize a right to continue a given use of real property if such use is
already in existence at the time of the enactment of a land use regulation forbidding or
restricting the land use in question”); State v. Thomasson, 378 P.2d 441, 443 (Wash. 1963)
(“[W]e have recognized that property owners are not without some rights in the continuance of a legitimate business on their property despite a change in zoning.”).
33 See, e.g., City of Annapolis v. Waterman, 745 A.2d 1000, 1015 (Md. 2000) (“[I]nsofar
as the ordinance required the reduction, removal, or destruction of existing property at the
owners’ expense, it was a taking . . . .”); see also Terminals Equip. Co. v. City & County of
San Francisco, 270 Cal. Rptr. 329, 335 (Cal. Ct. App. 1990) (rejecting takings claim because
regulation was “only operating to restrict new uses of the subject [p]roperty, not to eliminate existing ones”); Taylor v. Zoning Bd. of Appeals, 783 A.2d 526, 531 (Conn. App. Ct.
2001) (“The right to continue an established nonconforming use of one’s property is
securely grounded, both in statutes and in previous decisions of this court and our Supreme
Court.”).
34 Penn Cent. Transp. Co. v. City of New York, 438 U.S. 104, 124 (1978).
35 Id. at 136.
36 See, e.g., Wensmann Realty, Inc. v. City of Eagan, 734 N.W.2d 623, 638 (Minn. 2007)
(holding that owner of option to purchase golf course had no reasonable investmentbacked expectations to support claim that city’s denial of amendment to comprehensive
plan to permit residential development of property constituted taking); Consumers Union
of U.S., Inc. v. State, 840 N.E.2d 68, 103 (N.Y. 2005) (“The word ‘investment’ may seem
awkward in discussing the expectations of a not-for-profit entity, but I think the meaning
of ‘investment-backed expectations’ in this context is simply [claimant’s] reasonable expectations as to the future use of [its] property.”). The evolution of Penn Central’s
investment-backed expectations prong is discussed in Daniel R. Mandelker, The Notice
Rule in Investment-Backed Expectations, in TAKING SIDES ON TAKINGS ISSUES: PUBLIC
AND PRIVATE PERSPECTIVES 21, 21–29 (Thomas E. Roberts ed., 2002).
37 Penn Central, 438 U.S. at 124 (describing multifactor test for takings liability).
1230
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
property remains, however, eliminating an existing use should not
necessarily trigger takings liability. More problematically, Penn
Central postdates the development of zoning and the protection of
existing uses by almost half a century. If anything, Penn Central
uncritically incorporates the same underlying assumption about
existing uses without providing any strong constitutional or theoretical
underpinnings. The Due Process Clause, the other primary constitutional source for protecting property, fares no better.38
Part III examines possible normative justifications for existing
use protection and finds them all lacking. Existing uses seem to
deserve special protection, but it is surprisingly difficult to articulate
exactly why. At first blush, obvious possible rationales include a general sense of fairness and the desire to protect property owners’ reliance on existing regulations. Looking more carefully, however,
neither rationale clearly justifies distinguishing between existing uses
and prospective future uses. True, it seems unfair for a government to
change the regulatory rules in the middle of the game, thereby interfering with owners’ reasonable reliance on preexisting rules. But
these fairness concerns are at issue whenever the government regulates—whether the regulation affects existing uses or not. The expectations of an owner who purchased undeveloped property planning to
build a large development are also undermined by a downzoning of
the property, even though it affects only her intended use and not an
existing one. Arguments in favor of protecting existing uses assume
that there is something different, truly different, about the unfairness
associated with regulating existing as opposed to future uses. But
merely invoking fairness or reliance does not explain what this difference is.
Other superficially appealing justifications for existing use protection also lose their sheen on closer inspection. Extent of loss, for
example, does not turn on the existence of a use. Compare the harm
to a developer prevented from building a new condominium development with the harm of eliminating a decrepit shack in the woods. The
loss from the former may be in the millions of dollars, the loss from
the latter in the thousands, and yet it is the latter that the law more
vigorously protects. Other possible justifications are similarly
unconvincing.39
Finally, Part IV critiques a broader functional justification: that
the protection of existing uses is useful as an easy-to-administer
38
For a discussion of the Due Process Clause, see infra Part II.C.
The other possible justifications for existing use protection considered in this Article
include the endowment effect, political economy, waste prevention, owner expectations,
and the stability of the status quo. See infra Part III.
39
November 2009]
EXISTING USES
1231
bright-line rule for property protection. However, even deciding what
counts as an existing use turns out to be no easy task, undercutting
this administrability justification. There are, moreover, offsetting
pressures that militate against protecting existing uses. In addition to
general concerns about limiting the government’s ability to regulate,
the legal transitions literature suggests that protecting existing uses
can lead to inefficient investments in property by inducing owners to
race to lock in a use in anticipation of a rule change.40 While this is a
familiar insight from the transitions literature, it has particular force in
the land use context precisely because of the previously unexamined
ubiquity of existing use protection.
As a result, this Article concludes that courts should not protect
existing uses any differently than they protect prospective future uses.
The categorical difference in the treatment of existing and prospective
future uses is not compelled by either constitutional or normative considerations, and it comes with substantial costs.
Legislatures or policymakers may well have good reason to protect existing uses in any particular situation. Concerns about transition costs, fairness, political opposition, and the like may well cut in
favor of some existing use protection. It may often make good sense
for land use regulations to treat existing uses differently,
grandfathering them into zoning regulations and offering compensation for their elimination. But this will not always be true, and where
it is not, legislatures should be free to act, or at least more free than
courts’ current assumptions about existing use protection will allow.
Quite simply, existing uses should not receive the kind of categorical
protection from courts that property and land use law currently offer.
This is not to say that existing uses should receive no protection
at all. Instead, this is an argument for symmetry: that existing and
prospective uses should be subject to the same takings and due pro40 See, e.g., David A. Dana, Natural Preservation and the Race To Develop, 143 U. PA.
L. REV. 655, 677–95 (1995) (describing how natural preservation regulation encourages
development); Kaplow, supra note 9, at 527–31 (analyzing risks and incentives created by
government regulation); see also SHAVIRO, supra note 9, at 104–15 (discussing norm
against nominally retroactive taxes and implications of this norm); Abraham Bell &
Gideon Parchomovsky, Givings, 111 YALE L.J. 547, 582–84 (2001) (illustrating effect of
future regulatory takings on investment decisions); Saul Levmore, Changes, Anticipations,
Reparations, 99 COLUM. L. REV. 1657 (1999) (exploring incentives created by changes in
law); Kyle D. Logue, Legal Transitions, Rational Expectations, and Legal Process, 13 J.
CONTEMP. LEGAL ISSUES 211, 216–18 (2003) (arguing for transition policies for regulatory
takings of private property as type of “government-provided insurance”). For a leading
account of moral hazard, see generally Tom Baker, On the Genealogy of Moral Hazard, 75
TEX. L. REV. 237 (1996). Moral hazard in the land use context is considered in detail in
Part IV.A, infra.
1232
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
cess analyses. The Constitution, in other words, should not distinguish
between the two.
I
EXISTING USES
IN THE
LAW
The protection of existing uses is on display across a wide crosssection of law. From important state land use doctrines to various
statutes, the law contains a strong background rule protecting existing
uses from government regulation. Many doctrines explicitly invoke
constitutional restrictions on the government’s ability to regulate
existing uses, although the exact source and content of the constitutional protection are unclear.41 This Part explores significant land use
concepts and doctrines—retroactive zoning, nonconforming uses and
amortization, vested rights, and nuisance exceptions—that either
explicitly protect or implicitly assume the protection of existing uses.42
A. Retroactive Zoning
Zoning is perhaps the most obvious context in which the problem
of existing uses regularly arises. The issue can be traced directly back
to the Standard State Zoning Enabling Act (SZEA).43 Promulgated
in the 1920s, the SZEA represented an effort by the Commerce
Department to provide a blueprint for states to grant municipalities
the power to zone.44 It was drafted as model legislation, intended to
be adopted in its entirety by individual states.45 By the 1930s, a
41
See infra Part II.
The list is not exhaustive. The protection of existing uses is found in other areas too.
According to a leading casebook, for example, local governments may be more likely to
grant variances for structures that already exist than for those yet to be built. See
ELLICKSON & BEEN, supra note 6, at 293 (“[P]eople who built first and sought variances
later were more successful in obtaining variances than those who proceeded lawfully.”).
43 STANDARD STATE ZONING ENABLING ACT (Advisory Comm. on Zoning, U.S. Dep’t
of Commerce 1926), available at http://myapa.planning.org/growingsmart/pdf/SZEnabling
Act1926.pdf [hereinafter SZEA]; see Douglas W. Kmiec, Deregulating Land Use: An
Alternative Free Enterprise Development System, 130 U. PA. L. REV. 28, 58 (1981) (finding
that “most zoning ordinances” either completely or partially exempt “land uses that predate the ordinance” and remarking that “there was some feeling on the part of the drafters
of the SZEA that any attempt to apply zoning to existing land uses and structures without
compensation would have been found unconstitutional”). The issue goes back further than
that. The famous case Hadacheck v. Sebastian, involving a municipality eliminating an
existing use, dates back to 1915. 239 U.S. 394 (1915).
44 Stuart Meck, The Legislative Requirement that Zoning and Land Use Controls Be
Consistent with an Independently Adopted Local Comprehensive Plan: A Model Statute, 3
WASH. U. J.L. & POL’Y 295, 297 (2000).
45 The SZEA itself provides:
3. Modify this standard act as little as possible.—It was prepared with a full
knowledge of the decisions of the courts in every case in which zoning acts
have been under review, and has been carefully checked with reference to sub42
November 2009]
EXISTING USES
1233
majority of states had adopted some form of the SZEA, and in the
subsequent decades almost every state followed suit.46
The SZEA recognized and wrestled with the problem of existing
uses and ultimately allowed zoning to apply to them. Paragraph 9 of
the explanatory notes to the SZEA, with an illuminating double negative, provides:
No declaration that act is not retroactive.—Some laws contain a provision to the effect that “the powers by this act conferred shall not
be exercised so as to deprive the owner of any existing property of
its use or maintenance for the purpose to which it is then lawfully
devoted.” While the almost universal practice is to make zoning
ordinances nonretroactive, it is recognized that there may arise local
conditions of a peculiar character that make it necessary and desirable to deal with some isolated case by means of a retroactive provision affecting that case only. For this reason it does not seem wise
to debar the local legislative body from dealing with such a
situation.47
This provision acknowledges that the overwhelming convention—
indeed, the “almost universal practice”—is to have zoning apply only
prospectively, that is, only to future uses but not to existing ones. The
SZEA therefore recognizes that there is something special about
existing uses but it does not exempt them from zoning. It also reflects
an implicit assumption that the Constitution does not compel their
protection either.
Not all states followed the SZEA; some state zoning enabling acts
explicitly forbade application of zoning to existing uses.48 Meanwhile,
even in states that followed the SZEA, courts struck down as unconstitutional some zoning ordinances that applied to existing uses.49 In
sequent decisions. A safe course to follow is to make only those changes necessary to have the act conform to local legislative customs and modes of
expression.
SZEA, supra note 43, at 1.
46 Carol M. Rose, Planning and Dealing: Piecemeal Land Controls as a Problem of
Local Legitimacy, 71 CAL. L. REV. 837, 848 & n.29 (1983).
47 SZEA, supra note 43, at 2.
48 The zoning enabling acts of Illinois, Kansas, Massachusetts, New Hampshire, Ohio,
and Wisconsin provided that zoning could not apply to existing uses. See Comment, Retroactive Zoning Ordinances, 39 YALE L.J. 735, 735 & n.6 (1930) (citing statutes). But see
MARC A. WEISS, THE RISE OF THE COMMUNITY BUILDERS: THE AMERICAN REAL
ESTATE INDUSTRY AND URBAN LAND PLANNING 81 (1987) (describing Los Angeles ordinance predating zoning that permitted application of land use restrictions to existing
businesses).
49 See, e.g., Jones v. City of Los Angeles, 295 P. 14, 18–19, 22 (Cal. 1930) (striking down
ordinance and citing, inter alia, A.C. Blumenthal & Co. v. Cryer, 236 P. 216 (Cal. Dist. Ct.
App. 1925), Frank J. Durkin Lumber Co. v. Fitzsimmons, 147 A. 555, 558 (N.J. 1929),
Alfred Bettman, Constitutionality of Zoning, 37 HARV. L. REV. 834, 853 (1924), and J.S.
1234
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
Jones v. City of Los Angeles, the California Supreme Court heard a
challenge to a provision in the Los Angeles zoning ordinance that
made it illegal to operate or maintain a sanitarium outside of certain
limited areas.50 The ordinance did not distinguish between existing
and future sanitariums.51 After finding the zoning ordinance constitutional as applied prospectively, the court asked whether the same reasoning could justify the destruction of existing businesses.52 It held
that “[o]nly a paramount and compelling public necessity could sanction so extraordinary an interference with useful business.”53 Finding
none in that case, the court struck down the zoning ordinance as
applied to existing sanitariums.54 As another court later articulated,
“It is fundamental that a zoning regulation may not operate retroactively to deprive a property owner of his previously vested rights, that
is, a zoning regulation cannot deprive the owner of a use to which his
property was put before the zoning regulation became effective.”55
Commentators from the 1920s and 1930s, writing about the
SZEA, also doubted the validity of a zoning ordinance that applied to
existing uses.56 A comment in the Yale Law Journal from 1930 noted
that “it has been generally assumed that any attempt to make zoning
ordinances retroactive would meet with the opposition of the courts
and might result in their declaring the ordinances as a whole unconstitutional.”57 As is typical, however, the comment did not explain the
nature or source of the constitutional limitation.
Young, City Planning and Restrictions on the Use of Property, 9 MINN. L. REV. 593, 628
(1925)).
50 295 P. 14 (1930).
51 Id. at 15.
52 Id. at 17.
53 Id. at 19.
54 Id. at 22.
55 Bd. of County Comm’rs v. Petsch, 109 N.W.2d 388, 390 (Neb. 1961); see also Wayne
McCormack, Lochner, Liberty, Property, and Human Rights, 1 N.Y.U. J.L. & LIBERTY
432, 459–60 & n.153 (2005) (finding in Village of Euclid v. Ambler Realty Co., 272 U.S. 365
(1926), “some slight suggestion that failure to grandfather existing uses when enacting a
new zoning provision might be a problem”).
56 See, e.g., Bettman, supra note 49, at 853 (noting that application of zoning to existing
uses would present same constitutional problem as retroactive laws more generally);
Young, supra note 49, at 628 (“Retroactive zoning is not to be recommended except in
very unusual cases [when] public protection imperatively demands it.”).
57 Comment, supra note 48, at 737; see also Newman F. Baker, Zoning Legislation, 11
CORNELL L.Q. 164, 174–75 (1926) (“The Standard Act does not declare that the zoning
ordinances shall not be retroactive. . . . [B]ut the practice would cause many cases of individual hardship if tried in a comprehensively zoned city which would result in raising the
issue of the constitutionality of zoning in general.”).
November 2009]
EXISTING USES
1235
B. Prior Nonconforming Uses and Amortization
Since the promulgation of the SZEA, rules have developed
around the protection of prior nonconforming uses—that is, preexisting uses that are not in conformity with new land use regulations.58
These separate rules and doctrines governing the treatment of prior
nonconforming uses further reveal the law’s special solicitude toward
existing uses.
Initially, the SZEA contemplated that prior nonconforming uses
would simply die out over time.59 The assumption was that either
their incompatibility with surrounding uses would make them economically unsustainable or they would be lost to fire, abandonment,
or some other event, without government interference.60 That
assumption has proven false. For some uses, their very nonconformity
is what makes them particularly valuable—in effect assuring their survival.61 In response, local governments have sought proactive ways,
short of outright condemnation, to speed the demise of noncon58 See Eric J. Strauss & Mary M. Giese, Elimination of Nonconformities: The Case of
Voluntary Discontinuance, 25 URB. LAW. 159, 160 (1993) (defining nonconforming use as
“the lawful use of a building or premises, existing at the time of the adoption or amendment of a zoning ordinance, although such use does not conform to the provisions of the
ordinance” (quoting WIS. STAT. § 62.23(7)(h) (1988))). Courts and particularly commentators distinguish between nonconforming structures and nonconforming uses, although
essentially the same reasoning applies to both. See id. at 161.
59 See City of Los Angeles v. A.I. Gage, 274 P.2d 34, 40 (Cal. Dist. Ct. App. 1954)
(“Until recently zoning ordinances have made no provision for any systematic and comprehensive elimination of the nonconforming use. The expectation seems to have been that
existing nonconforming uses would be of little consequence and that they would eventually
disappear. The contrary appears to be the case.” (internal citations omitted)); Osborne M.
Reynolds, Jr., The Reasonableness of Amortization Periods for Nonconforming Uses—
Balancing the Private Interest and the Public Welfare, 34 WASH. U. J. URB. & CONTEMP. L.
99, 99–100 (1988) (noting that courts also believed that nonconforming uses would die out
and that courts have often assumed, where ordinance did not specify, that ordinance did
not apply retroactively).
60 See Varadarajan, supra note 13, at 2556 (“Zoning regulators believed that these
restrictions would cause the gradual disappearance of such nonconforming uses.”); see also
Huntington Props., LLC v. Currituck County, 569 S.E.2d 695, 700 (N.C. Ct. App. 2002)
(“Moreover, ‘non-conforming uses are not favored by the law. Most zoning schemes
foresee elimination of non-conforming uses either by amortization, or attrition or other
means.’” (quoting CG & T Corp. v. Bd. of Adjustment, 411 S.E.2d 655, 659–60 (N.C. Ct.
App. 1992))). This view traditionally justifies laws limiting improvements or expansions of
nonconforming uses. See, e.g., County Council v. E.L. Gardner, Inc., 443 A.2d 114, 119
(Md. 1982) (recognizing that “the purpose of such restrictions is to achieve the ultimate
elimination of nonconforming uses through economic attrition and physical
obsolescence”).
61 See Reynolds, supra note 59, at 109 (“[N]onconforming uses thrived due to the protection from new competition given them by the zoning laws.”); Strauss & Giese, supra
note 58, at 163 (“Unfortunately, nonconforming uses were not phased out because the
restriction on the development of similar uses in the area created a virtual monopoly,
which allowed the nonconforming use to flourish.”).
1236
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
forming uses.62 These efforts have spawned robust and varied rules
about when a property owner is entitled to expand, adapt, or rebuild
such a use.63 Also, in many states local governments can affirmatively
eliminate prior nonconforming uses without paying compensation by
allowing an affected property owner to continue her use long enough
to amortize her investment.64 Fundamentally, amortization is “a technique for the removal of non-conforming uses after the value of a
non-conforming use has been recovered—or amortized—over a
period of time. . . . Since the value of the use has been amortized, no
compensation is payable after the expiration of the period.”65
The use of amortization strikes many as unfair, and courts disagree about its constitutional basis and its limits. Some state courts
have held that amortization in lieu of compensation is simply impermissible.66 Other courts have held that amortization is a constitutional means of eliminating prior nonconforming uses, so long as the
amortization period is sufficiently long.67 Much has been written
62 These include precluding any expansion of a nonconforming use and terminating a
nonconforming use if the use is abandoned, changed, or destroyed. See Craig A. Peterson
& Claire McCarthy, Amortization of Legal Land Use Nonconformities as Regulatory Takings: An Uncertain Future, 35 WASH. U. J. URB. & CONTEMP. L. 37, 39–41 (1989) (summarizing early approaches to eliminating existing uses); see also Reynolds, supra note 59, at
101–04 (describing limitations on growth and repair); Strauss & Giese, supra note 58, at
163–67 (summarizing all available methods).
63 For a summary and analysis of these rules, see Eunice A. Eichelberger, Annotation,
Alteration, Extension, Reconstruction, or Repair of Nonconforming Structure or Structure
Devoted to Nonconforming Use as Violation of Zoning Ordinance, 63 A.L.R. 4th 275
(1988).
64 See Peterson & McCarthy, supra note 62, at 37 (“[S]tate courts have generally
upheld amortization provisions since the 1950s.”); Reynolds, supra note 59, at 109 (“[M]ost
courts held that amortization provisions are valid if they are reasonable in nature. This is
currently the majority view in America.”). The actual use of amortization may be quite
limited, however. See Margaret Collins, Methods of Determining Amortization Periods for
Non-conforming Uses, 3 WASH. U. J.L. & POL’Y 215, 216 (2000) (“A survey of 489 cities
showed that, although planners in 159 cities had access to amortization programs, only 27
cities had actually used them.”).
65 Collins, supra note 64, at 216. Amortization is related to the depreciation of an asset
for tax purposes. See id. at 223–24 (explaining various depreciation methods).
66 See, e.g., City of Oakbrook Terrace v. Suburban Bank & Trust Co., 845 N.E.2d 1000,
1011 (Ill. App. Ct. 2006) (holding that municipal ordinance providing amortization period
for nonconforming signs violated State Eminent Domain Act, which required just compensation); Hoffmann v. Kinealy, 389 S.W.2d 745, 753 (Mo. 1965) (“Of course, every comprehensive zoning ordinance limits and thereby regulates the use of property prospectively.
But we cannot embrace the doctrine espoused by advocates of the amortization technique
that there is no material distinction between regulating the future use of property and
terminating pre-existing lawful nonconforming uses.”); PA Nw. Distribs., Inc. v. Zoning
Hearing Bd., 584 A.2d 1372, 1376 (Pa. 1991) (holding that amortization provisions are
unconstitutional because they take property without just compensation).
67 See, e.g., Bd. of Zoning Appeals v. Leisz, 702 N.E.2d 1026, 1032 (Ind. 1998) (holding
that amortization provisions that require owner to discontinue nonconforming use after
November 2009]
EXISTING USES
1237
about the amount of time that governments must give property
owners before an existing use can be eliminated and about when
amortization statutes can and should be applied.68 In general, courts
considering the elimination of a prior nonconforming use weigh the
harm to the property owner against the benefit to the public.69 The
test sounds primarily in due process.70
These doctrinal nuances can be set aside for now, however, in
order to focus on the underlying assumptions on which existing use
doctrine rests. The background rule, even in jurisdictions that allow
amortization, is that governments cannot eliminate existing uses
certain period of time are not per se unconstitutional); Trip Assocs. v. Mayor and City
Council of Baltimore, 898 A.2d 449, 457 (Md. 2006) (“So long as it provides for a reasonable relationship between the amortization and the nature of the nonconforming use, an
ordinance prescribing . . . amortization is not unconstitutional.”); Red Roof Inns, Inc. v.
City of Ridgeland, 797 So. 2d 898, 902 (Miss. 2001) (“The weight of authority supports the
conclusion that a reasonable amortization provision would not be unconstitutional.” (citation omitted)). The inquiry into duration is also a source of controversy, with some courts
holding that amortization must be sufficiently long to allow the value of the buildings on
the property to depreciate fully, and others holding that the adequacy of the duration must
be evaluated by comparing the burden to the property owner with the burden to the public
of allowing the use to continue. Compare Lone v. Montgomery County, 584 A.2d 142, 153
(Md. Ct. Spec. App. 1991) (“We hold that the ten-year amortization period in the case at
bar is reasonable because within that time, the multi-family housing owners could recoup
any lost investment . . . .”), with Modjeska Sign Studios, Inc. v. Berle, 373 N.E.2d 255, 262
(N.Y. 1977) (“If an owner can show that the loss he suffers as a result of the removal of a
nonconforming use at the expiration of an amortization period is so substantial that it
outweighs the public benefit gained by the legislation, then the amortization period must
be held unreasonable.”).
68 Cf. Reynolds, supra note 59, at 108 (describing constitutional concern about duration
of amortization period). See generally Collins, supra note 64.
69 See Modjeska Sign Studios, 373 N.E.2d at 262 (“In essence, however, we believe the
critical question which must be asked is whether the public gain achieved by the exercise of
the police power outweighs the private loss suffered by owners of nonconforming uses.”);
Collins, supra note 64, at 217 (“The process of determining amortization periods is . . . a
‘balancing test’ weighing the private cost against the public gain.”). But see Peterson &
McCarthy, supra note 62, at 72–79 (applying Penn Central test to as applied challenge to
amortization provisions).
70 See Lingle v. Chevron, 544 U.S. 528, 540 (2005) (describing due process test). In fact,
if the test for amortization of prior existing uses is not more protective than substantive
due process, it threatens to become superfluous. In a dissenting opinion, a judge on the
Washington Court of Appeals recognized this point:
The majority’s flaw is its failure to make the distinction between continuation
of a nonconforming use which is exempt from police power regulation on the
one hand, and imposition of the police power without exemption, subject only
to the usual requirements of due process, on the other. If the latter be the rule,
the nonconforming use doctrine is robbed of its reason for existence, and is no
more than the usual due process test.
Rhod-A-Zalea & 35th, Inc. v. Snohomish County, 959 P.2d 1024, 1036 (Wash. 1988) (en
banc) (Sanders, J., dissenting).
1238
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
immediately.71 Even where amortization is permissible, then, it is a
narrow and constrained exception to the general rule that the government cannot eliminate existing uses without paying compensation.
The source of that underlying and fundamental restriction, however, is
left unstated.
C. The Vested Rights Doctrine
Outside of the zoning context, one of the easiest places to discern
the special protection of existing uses is in the vested rights doctrine,
which defines when a property owner has taken sufficient steps to lock
in existing land use regulations.72 The doctrine is usually implicated
when a property owner has begun but not yet completed some project
before the government changes the applicable regulations. In its most
general form, the vested rights doctrine defines when, and under what
circumstances, an incomplete project can count as an existing use.73
The doctrinal details determining when a right vests are the subject of
frequent commentary.74 It is the implicit assumption motivating the
doctrine that is of particular interest here, however, not the niceties of
the doctrine itself.
The vested rights doctrine assumes that if a right has vested—that
is, if a project is sufficiently far along—then it is entitled to protection
from subsequently enacted land use regulations (at least those not
falling within the other exceptions described below).75 Under the
71 Richard B. Cunningham & David H. Kremer, Vested Rights, Estoppel, and the Land
Development Process, 29 HASTINGS L.J. 623, 660–61 (1978) (“Early cases and treatises
contain numerous admonitions based on reasons at once philosophical, practical, and legal
that zoning must be prospective in nature. The assumption has not changed dramatically
and continues to be of paramount importance in many modern land regulatory statutes.”);
Peterson & McCarthy, supra note 62, at 39 (finding that “pre-existing uses and aspects of
development are always to some extent ‘grandfathered’”); Reynolds, supra note 59, at 104
(“Supporters of [amortization] have generally agreed that constitutional limitations, or
considerations of fairness require that existing uses be allowed to continue until the user
has had a reasonable opportunity to amortize his investment.” (footnotes omitted)
(internal quotation marks omitted)).
72 See Cunningham & Kremer, supra note 71, at 625–26 (summarizing vested rights
doctrine).
73 For a typical hypothetical raising this problem, see CHARLES L. SIEMON ET AL.,
VESTED RIGHTS 2 (1982) (introducing book with stylized hypothetical of developer
affected by zoning regulations passed after significant investment).
74 See generally Gregory Overstreet & Diana M. Kirchheim, The Quest for the Best Test
To Vest: Washington’s Vested Rights Doctrine Beats the Rest, 23 SEATTLE U. L. REV. 1043,
1069–71 (2000) (endorsing Washington State’s vested rights rule in part because it provides
“date certain” when development rights will vest); E.A. Prichard & Gregory A. Riegle,
Searching for Certainty: Virginia’s Evolutionary Approach to Vested Rights, 7 GEO. MASON
L. REV. 983 (1999) (describing complex history of vested rights issue in Virginia).
75 See, e.g., Trans-Oceanic Oil Corp. v. City of Santa Barbara, 194 P.2d 148, 152 (Cal.
Dist. Ct. App. 1948) (holding that valid permit “ripens into a vested property right which
November 2009]
EXISTING USES
1239
vested rights doctrine, then, property rights are subject to a kind of
tipping point. Before a right vests, property receives one level of protection. But as soon as an owner has done enough to establish a particular use, the property becomes entitled to much greater protection.
Consider, for example, Prince George’s County v. Sunrise Development Ltd. Partnership. 76 There, the developer of a proposed
twelve-story apartment building sued to prevent a downzoning. The
parties all agreed that if the developer had actually begun construction of the building, the downzoning could not have applied to the
property.77 The question in the case was whether pouring a footing
for a single column to support a proposed portico counted as a sufficient start to the building process to vest the development rights.78
The court described how the footing had been covered with plastic
and straw and so was not visible to building inspectors,79 how a teninch by ten-inch by one-inch steel plate had not been affixed to the
top of the footing,80 and how the footing ultimately occupied only four
square feet on the site.81 The Maryland Court of Appeals ultimately
reversed the lower court and held that this footing was not enough to
vest development rights.82 Had the construction progressed past that
point, perhaps by building enough above ground to be visible to
building inspectors—then the development rights would have vested
and the government suddenly would have been powerless to
downzone the property without compensation.
The vested rights doctrine can therefore best be explained as
incorporating an assumption that existing uses receive special protecmay not be taken from him against his will other than by proceedings in eminent domain
with the payment of just compensation”). According to Black’s Law Dictionary, a vested
right is “[a] right that so completely and definitely belongs to a person that it cannot be
impaired or taken away without the person’s consent.” BLACK’S LAW DICTIONARY 1349
(8th ed. 2004).
76 623 A.2d 1296 (Md. 1993).
77 Id. at 1301.
78 Id. at 1297 (“The developer’s vested rights argument ultimately rests on a single
footing, one that was placed in the ground . . . for a proposed column at a proposed outside
corner of a proposed portico.”).
79 Id. at 1298–99.
80 Id. at 1298–99 n.3.
81 Id. at 1302.
82 Specifically, the court reinstated an administrative board’s finding that development
rights had not vested, holding:
From the standpoint of a member of the general public who is either viewing
the property from its boundaries or is consensually on the property, the footing
is not so clearly the commencement of construction as to render the Board’s
finding to the contrary arbitrary, capricious or without substantial evidence on
the entire record.
Id. at 1305.
1240
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
tion from government regulation. The focus of the vested rights
inquiry is on whether the property owner has done enough—i.e., has
taken sufficient affirmative steps—to be entitled to put her property
to an intended use.83 Once a use has been legally established, it is
entitled to protection as an existing use. In other words, the line
between a vested right and a mere expectancy accords with the line
between an existing and a prospective future use.
D. The Nuisance Exception
A broad but problematic exception to the protection of existing
uses is the so-called “nuisance exception” to the Takings Clause.84
Applying the nuisance exception, the government can regulate away a
hazardous or injurious activity without paying compensation. Usually
left unstated, however, is the implicit assumption that non-harmful
uses cannot be eliminated without compensation. Framed as an
exception to takings liability, the nuisance inquiry again assumes and
reinforces the background rule that existing uses cannot be eliminated
unless they are nuisances.
Most famously, in Hadacheck v. Sebastian, the Supreme Court
upheld a statute that eliminated an existing brickyard from a residential neighborhood.85 Similarly, two early twentieth-century cases from
Louisiana upheld a zoning ordinance that required existing businesses
to liquidate or move out of residential neighborhoods within one year
on grounds of harm prevention.86 Colorfully, in 1919 the Supreme
Court of California upheld an ordinance that prohibited keeping
83 See, e.g., John J. Delaney, Vesting Verities and the Development Chronology: A
Gaping Disconnect?, 3 WASH. U. J.L. & POL’Y 603, 607 (2000) (“At least 30 state courts
have used the issuance of a building permit as the principal benchmark for [vested rights],
but virtually all of these courts also require that other actions be taken in reliance upon the
permit, such as construction or expenditure of funds to implement the permit.”).
84 See Mark Fenster, The Takings Clause, Version 2005: The Legal Process of Constitutional Property Rights, 9 U. PA. J. CONST. L. 667, 674 (2007) (“[T]he Supreme Court has
typically, although not universally, allowed government to regulate broadly against nuisance activities and thereby lower private property value without compensation, especially
where the regulation provided reciprocal benefits to the affected property owner.”);
Christine A. Klein, The New Nuisance: An Antidote to Wetland Loss, Sprawl, and Global
Warming, 48 B.C. L. REV. 1155, 1195 (2007) (“In theory, traditional takings law has long
recognized a nuisance exception under which landowners are not entitled to compensation
when they are precluded from using their land to create a nuisance.”).
85 239 U.S. 394 (1915). Mugler v. Kansas, 123 U.S. 623 (1887), is similar, rejecting a
constitutional challenge to prohibition in Kansas on the basis of a harm-prevention rationale. More perniciously, in 1911 the California Supreme Court upheld legislation eliminating 110 Chinese laundries on finding that they caused a harm. See Ex Parte Quong Wo,
118 P. 714 (Cal. 1911), discussed in Young, supra note 49, at 627.
86 Comment, supra note 48, at 736–37 (citing State v. McDonald, 121 So. 613 (La.
1929), and State ex rel Dema Realty Co. v. Jacoby, 123 So. 314 (La. 1929)).
November 2009]
EXISTING USES
1241
mules in residential neighborhoods.87 Finding the ordinance reasonable, that court wrote:
[W]e know of no heaven-sent maxim to invent a silencer for this
brute, that one beholding him, neck outstretched and jaws distended wide, could persuade himself that he but heard from the
depths of the beast’s crimson coated cavern . . . a sound so fine
there’s nothing lives ‘[t]wixt it and silence.88
Similar, if less colorful, cases are now legion.89
Instead of repudiating robust protection for existing uses, these
cases are part of a distinct and more limited doctrine that permits
eliminating existing uses for nuisance or harm prevention.90 In other
words, harm prevention is an exception to the implicit background
rule that the government cannot simply regulate away an existing use.
Where a regulation eliminating an existing use is not preventing a
harm, courts have held the regulation invalid in the absence of compensation.91 Of course, as many have observed, it is hard to find a
limiting principle that distinguishes between a regulation preventing a
harm and a regulation conferring a benefit,92 but that such a line even
87 Boyd v. City of Sierra Madre, 183 P. 230 (Cal. 1919), quoted in Young, supra note 49,
at 616.
88 Id. at 232 (ellipsis in original) (internal quotation marks omitted).
89 See, e.g., Goldblatt v. Hempstead, 369 U.S. 590 (1962) (upholding regulation banning
excavations below water table despite effect of eliminating gravel mining operation in existence for thirty years); Dep’t of Agric. & Consumer Servs. v. Polk, 568 So. 2d 35 (Fla. 1990)
(holding that destruction of trees exhibiting bacterial disease did not constitute taking);
Zeman v. City of Minneapolis, 552 N.W.2d 548, 555 (Minn. 1996) (denying compensation
where landlord’s rental dwelling license was revoked after tenants received multiple disorderly conduct citations because ordinance “serves a public harm prevention purpose”).
90 See Young, supra note 49, at 627 (“Some cities have made their ordinances retroactive for industries [that are] more or less offensive.”); Comment, supra note 48, at 737 (“It
is a common occurrence, where there are zoning ordinances in effect, for the exclusions to
begin with nuisances and near-nuisances, over which municipalities have always had extensive control.”). Moreover, Hadacheck triggered a backlash against Los Angeles’s early
twentieth-century land use regulations that applied to existing uses: Following Hadacheck,
many land use statutes exempted existing uses. WEISS, supra note 48, at 86.
91 See, e.g., Sintra, Inc. v. City of Seattle, 829 P.2d 765, 773 (Wash. 1992)
(“[R]egulations which enhance public interests, and go beyond preventing harmful activity,
may constitute a taking. The regulatory scheme here goes beyond preventing harm.”); see
also Lucas v. S.C. Coastal Council, 505 U.S. 1003, 1028–30 (1992) (holding that regulations
which prohibit all economically beneficial use of land are unconstitutional without compensation to burdened property owners unless that regulation is proscribing use under
established property and nuisance principles); Robinson v. City of Seattle, 830 P.2d 318,
328 (Wash. 1992) (“[I]f the regulation . . . goes beyond mere harm prevention to require a
property owner to provide a public benefit, then that regulation is susceptible to a constitutional taking challenge.”).
92 See, e.g., Glynn S. Lunney, Jr., Responsibility, Causation, and the Harm-Benefit Line
in Takings Jurisprudence, 6 FORDHAM ENVTL. L. REV. 433, 475–78 (1995); see also Lucas,
505 U.S. at 1024 (observing that difference between preventing harm and conferring benefit is “often in the eye of the beholder”).
1242
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
needs to be drawn demonstrates that existing uses are normally
protected.
II
CONSTITUTIONAL PROTECTION
FOR
EXISTING USES
As Part I demonstrated, a strong current running through land
use law protects existing uses from being regulated away without compensation. This protection is usually based upon perceived constitutional requirements whose nature and content are defined only
vaguely. Looking closely at the supposed doctrinal underpinnings for
existing use protection reveals how startlingly weak they are. Courts
frequently invoke the Constitution as compelling a higher level of protection for existing uses but fail to articulate precisely why and how
this level of protection applies. Instead of a coherent set of doctrines
based on sound constitutional footing, existing use protection is generally an unexamined assumption that courts blindly follow.
Hansen Bros. Enterprises v. Board of Supervisors 93 is representative. In that case, the Supreme Court of California wrote:
Zoning ordinances and other land use regulations customarily
exempt existing uses to avoid questions as to the constitutionality of
their application to those uses. “The rights of users of property as
those rights existed at the time of the adoption of a zoning ordinance are well recognized and have always been protected.”94
In its general discussion of zoning’s application to existing uses, the
Hansen Brothers court cited Penn Central,95 the foundational modern
Takings Clause case, as well as Euclid v. Ambler Realty Co.,96 the case
that upheld zoning against a facial Due Process Clause challenge.97
The court, however, went no further either in analyzing the Takings
Clause or the Due Process Clause or in explaining why or how either
Clause protects existing uses. This is quite typical.98
It is therefore no simple task even to identify the supposed constitutional basis for the land use doctrines discussed in Part I. The next
93
907 P.2d 1324 (Cal. 1996).
Id. at 1335 (quoting Edmonds v. Los Angeles County, 255 P.2d 772, 777 (Cal. 1953)).
95 Id. (citing Penn Cent. Transp. Co. v. New York City, 438 U.S. 104, 125 (1978)).
96 Id. (citing Euclid v. Ambler Realty Co., 272 U.S. 365 (1926)).
97 Euclid, 272 U.S. at 397.
98 See, e.g., Calvert v. County of Yuba, 145 Cal. App. 4th 613, 623 (2006) (“In light of
the state and federal constitutional takings clauses, when zoning ordinances or similar land
use regulations are enacted, they customarily exempt existing land uses (or amortize them
over time) to avoid questions as to the constitutionality of their application to those
uses.”); Prince George’s County v. Sunrise Dev. Ltd. 623 A.2d 1296, 1303–04 (Md. 1993)
(stating, without elaboration, that vested rights doctrine has “constitutional foundation”).
For other examples, see supra notes 1, 32, and 33.
94
November 2009]
EXISTING USES
1243
section undertakes this necessary first step; the sections that follow
examine more fully how takings and due process law actually apply to
existing uses.
A. Constitutional Underpinnings
The constitutional basis for protecting existing uses is not at all
clear. Looking first at zoning, early cases challenging zoning ordinances were contemporaneous with Pennsylvania Coal Co. v. Mahon
(Penn Coal) but predate any conception of the Takings Clause as providing robust property protection from land use regulations.99 It is
therefore not surprising that the early limits on zoning power appear
to come from general limits on the police power and from the Due
Process Clause more broadly.100 Courts often demanded an important or compelling government reason for zoning existing uses and
stood ready to invalidate zoning ordinances that did not have one.101
As zoning authorities have grown more sophisticated, however, they
have come to employ more targeted techniques to eliminate existing
uses. These techniques implicate a more specific but still unsettled
constitutional inquiry.
Courts’ treatment of amortization statutes highlights the doctrinal
confusion surrounding existing uses generally and prior nonconforming uses in particular. A number of courts have held that an
amortization period is nothing but a deferred taking of property.102
They reason that eliminating an existing use immediately would be an
unconstitutional taking of property and that putting off eliminating
the use until tomorrow (or for a fixed amortization period) is no more
constitutional.103 Allowing a business or use to remain in existence
99 Penn Coal, the case that famously identified the possibility of regulatory takings, was
decided in 1922. Pa. Coal Co. v. Mahon, 260 U.S. 393 (1922). The Supreme Court did not
invalidate or even address the constitutionality of a land use regulation under the Takings
Clause until Penn Central in 1978. See 438 U.S. 104 (1978).
100 Bradley C. Karkkainen, The Police Power Revisited: Phantom Incorporation and the
Roots of the Takings “Muddle,” 90 MINN. L. REV. 826, 838–51 (2006).
101 See cases cited supra note 49.
102 E.g., Hoffmann v. Kinealy, 389 S.W.2d 745, 754–55 (Mo. 1965) (finding that elimination of prior nonconforming use of land for open storage of lumber and other building
materials and equipment would trigger takings liability under Missouri Constitution—“a
taking not to be justified as an exercise of the police power which is always subject to, and
may never transcend, constitutional rights and limitations”); PA Nw. Distribs., Inc. v.
Zoning Hearing Bd., 584 A.2d 1372, 1376 (Pa. 1991) (finding amortization of lawful preexisting use violative of Pennsylvania Constitution); cf. Collins, supra note 64, at 217 (“The
amortization technique, as applied to non-conforming uses, has been described as more of
a postponement than a solution.”); Reynolds, supra note 59, at 105 & n.24 (listing cases
where court did not distinguish between immediate and delayed cessation).
103 Hoffmann, 389 S.W.2d at 753 (stating that “no one has, as yet, been so brash as to
contend that . . . a pre-existing lawful nonconforming use might be terminated immedi-
1244
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
for a while—even for a few years—is no substitute for the compensation the Takings Clause requires.104
Other courts have struck down amortization provisions after
applying a substantive due process analysis.105 These courts reason
that land use regulations eliminating existing uses are generally arbitrary or unreasonable. As with all police powers, the zoning power is
limited by the Due Process Clause, and this limit requires an inquiry
into the public benefit and private harms of a zoning ordinance.106
Since, according to these courts, the private harm of eliminating an
existing use is very high, amortization is impermissible unless there is
an even more compelling public purpose.107
The majority of courts, however, have upheld amortization statutes—but only after applying a takings analysis. Some have reasoned
that the amortization period itself serves as a kind of implicit compensation sufficient to satisfy the Takings Clause’s requirement of just
compensation.108 Other courts have adopted similar reasoning,
holding that the revenue earned during the amortization period can
be included in regulatory takings analysis in order to decide whether
ately,” and further arguing that “the contrary is implicit in the amortization technique itself
which would validate a taking presently unconstitutional by the simple expedient of postponing such taking for a ‘reasonable’ time”).
104 See PA Nw. Distribs., 584 A.2d at 1376 (finding amortization period violative of
Pennsylvania Constitution); see also 83 AM. JUR. 2D Zoning and Planning § 622 & n.5
(2003) (collecting cases invalidating amortization rules under Takings Clause); Reynolds,
supra note 59, at 105–06 (same).
105 See, e.g., Concord Twp. v. Cornogg, 9 Pa. D. & C.2d 79, 86 (1956) (striking down
zoning ordinance with amortization provision of six months on basis of constitutional
ground “that no citizen can be deprived of his property without due process of law”); see
also Ailes v. Decatur County Area Planning Comm’n, 448 N.E.2d 1057, 1060 (Ind. 1983)
(“[A]n ordinance prohibiting any continuation of an existing lawful use within a zoned
area regardless of the length of time given to amortize that use is unconstitutional as the
taking of property without due process of law and an unreasonable exercise of the police
power.”), overruled by Bd. of Zoning Appeals v. Leisz, 702 N.E.2d 1026 (Ind. 1998).
106 See, e.g., Sun Oil Co. of Pa. v. City of Upper Arlington, 379 N.E.2d 266, 271 (Ohio
Ct. App. 1977) (determining that amortization provisions are valid only where they eliminate nuisances); see also Reynolds, supra note 59, at 108 (describing Due Process
protection).
107 See Modjeska Sign Studios, Inc. v. Berle, 373 N.E.2d 255, 262 (N.Y. 1977) (finding
amortization reasonable only when “the public gain achieved . . . outweighs the private loss
suffered by owners of nonconforming uses”); cf. Austin v. Older, 278 N.W. 727, 730 (Mich.
1938) (“An ordinance requiring an immediate cessation of a nonconforming use may be
held to be unconstitutional because it brings about a deprivation of property rights out of
proportion to the public benefit obtained . . . .”).
108 See, e.g., Outdoor Graphics, Inc. v. City of Burlington, 103 F.3d 690, 695 (8th Cir.
1996) (including benefits obtained during amortization period in regulatory takings analysis); see also Varadarajan, supra note 13, at 2573–76 (discussing amortization as substitute
for just compensation).
November 2009]
EXISTING USES
1245
the regulation has “go[ne] too far.”109 Still other courts have held that
amortization is actually an alternative to just compensation that nevertheless satisfies the Takings Clause.110 Ultimately, it is unclear
whether amortization is to be evaluated under the Takings Clause, the
Due Process Clause, or both.
This same bifurcated constitutional approach applies, with even
greater complexity, to the vested rights doctrine. Many courts locate
the basis for protection of vested rights in the Takings Clause.111 Confoundingly, the phrase “vested rights” is used in at least two distinct
ways.112 Some cases define vested rights as protected property interests and distinguish them from mere expectations that do not even
implicate the Takings Clause.113 Used this way, vested rights have a
limiting character; courts will not even apply takings analysis to rights
that are not vested. The term is, in essence, synonymous with a property right, making the word “vested” superfluous.114 Other cases,
109 See Naegele Outdoor Adver., Inc. v. City of Durham, 803 F. Supp. 1068, 1078
(M.D.N.C. 1992) (finding no taking because earnings during amortization period exceeded
investment in property).
110 E.g., Temple Baptist Church, Inc. v. City of Albuquerque, 646 P.2d 565, 572 (N.M.
1982) (finding amortization period satisfied requirement of just compensation). In theory,
one principal justification for amortization is that it is pegged to the useful economic life of
the underlying structure. If the value of an asset has fully depreciated for tax purposes,
then the courts will uphold its elimination without payment of just compensation. See
Collins, supra note 64, at 227.
111 E.g., Sand Key Assocs. v. Bd. of Trustees of Internal Improvement Trust Fund, 458
So. 2d 369, 370–71 (Fla. Dist. Ct. App. 1984) (“[P]roperty rights are vested and cannot be
taken away without just compensation.”); Bye v. Giarrusso, 321 So. 2d 58, 60 (La. Ct. App.
1975) (“‘[V]ested rights’ shall not be divested except for purposes of public utility and then
only on payment of just compensation.” (quoting Quartemont v. Aroyelles Parish Police
Jury, 228 So. 2d 199, 201 (La. Ct. App. 1969))); Padgett v. Mason County Zoning Comm’n,
Nos. 236458, 236459, 2003 WL 22902878, at *5 (Mich. Ct. App. Dec. 9, 2003) (“[T]he prohibition of taking property without just compensation extends only to vested rights in
property.”).
112 Others have noted the use of the phrase “vested rights” as a substitute for real analysis. See Cunningham & Kremer, supra note 71, at 628 (“One common approach is to
presume that a perfunctory reference to vested rights explains the legal theory involved
and the reason for its application to the facts of the case.”).
113 See, e.g., Shrader v. United States, 38 Fed. Cl. 788, 796–97 (1997) (“[R]egarding a
possible takings claim by Mr. Shrader, plaintiff’s interest[s] . . . were neither vested rights,
nor were they more than a ‘mere expectancy, and therefore not entitled to protection as a
property right.’” (quoting Alberico v. United States, 783 F.2d 1024, 1027 (Fed. Cir. 1986));
Att’y Gen. v. Mich. Pub. Serv. Comm’n, 642 N.W.2d 691, 699 (Mich. Ct. App. 2002) (“One
who asserts an uncompensated taking claim must first establish that a vested property right
is affected.” (quoting Fun ‘N Sun RV, Inc. v. Michigan, 527 N.W.2d 468, 478 (Mich.
1994))); Tracy v. City of Deshler, 568 N.W.2d 903, 907 (Neb. 1997) (contrasting “vested
property right” with “mere privilege”); see also Cunningham & Kremer, supra note 71, at
639–40 (distinguishing between privilege and right).
114 In this sense, the term “vested right” is synonymous with the term “protected property interest” and, as such, its use is entirely circular. See Cunningham & Kremer, supra
note 71, at 640 (“[‘Vested’] is frequently applied in a circular fashion to describe any
1246
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
however, define vested rights as rights that are immune from government regulation absent just compensation (or the application of one
of the other exceptions discussed above).115 Here, the phrase “vested
right” is a legal conclusion that the property right at issue is protected
from regulation unless compensation is paid.116 In other words,
finding a “vested right” is sometimes used to begin the takings inquiry
and other times to end it.117
This confusion about the meaning of “vested rights” exists in
uneasy tension with the vested rights doctrine and the Takings Clause.
It is simply not the case that the Takings Clause applies only to development projects that have progressed past a certain stage. Indeed, an
owner of undeveloped land who has made no effort to develop her
property may nevertheless have a takings claim if the government dramatically downzones the property and sharply reduces its market
value.118 On the other hand, finding that a vested right is protected
property under the Fifth Amendment would not seem to end the
inquiry of whether a particular government action is an unconstitu-
interest that is protected by the law. Hence, to assert that a person has a vested right is
merely to announce the conclusion that a court will protect that interest.”). Any property
right, not just a vested property right, is potentially protected by the Takings Clause. John
J. Delaney & Emily J. Vaias, Recognizing Vested Development Rights as Protected Property
in Fifth Amendment Due Process and Taking Claims, 49 WASH. U. J. URB. & CONTEMP. L.
27, 31 (1996) (“In fact, many state courts use the words ‘vested right’ and ‘property right’
interchangeably.”); cf. 26 AM. JUR. 2D Eminent Domain § 150 (2004) (“For purposes of
just compensation in an eminent-domain proceeding, ‘property’ includes every sort of
interest the citizen may possess.”).
115 See, e.g., Landgraf v. USA Film Prods., 511 U.S. 244, 266 (1994) (“The Fifth
Amendment’s Takings Clause prevents the Legislature (and other government actors)
from depriving private persons of vested property rights except for a ‘public use’ and upon
payment of ‘just compensation.’”); see also Kenneth R. Kupchak et al., Arrow of Time:
Vested Rights, Zoning Estoppel, and Development Agreements in Hawai’i, 27 U. HAW. L.
REV. 17, 38 (2004) (quoting County of Kauai v. Pac. Standard Life Ins. Co., 653 P.2d 766,
780 & n.129 (Haw. 1982) and describing it as “equat[ing] establishing a vested right with
establishing a regulatory taking”); Ann Woolhandler, Public Rights, Private Rights, and
Statutory Retroactivity, 94 GEO. L.J. 1015, 1023–24 (2006) (“The vested label . . . often
seems conclusory.”).
116 According to Cunningham and Kremer, “[t]he term instead should be recognized as
the end product of a process that weighs and analyzes a private interest to determine
whether it is of sufficient status to receive legal protection.” Cunningham & Kremer, supra
note 71, at 641.
117 For a historical explanation for the confusion about vested rights, see James L.
Kainen, The Historical Framework for Reviving Constitutional Protection for Property and
Contract Rights, 79 CORNELL L. REV. 87, 104–14 (1993).
118 See, e.g., Lucas v. S.C. Coastal Council, 505 U.S. 1003 (1992) (holding that regulation
prohibiting building on beachfront lot, thus depriving landowner of virtually all economically viable use, constitutes compensable government taking).
November 2009]
EXISTING USES
1247
tional taking. Yet courts consistently reason as though a vested right
simply cannot be regulated away.119
Other courts and commentators have located the protection for
vested rights in the Due Process Clause, invoking concerns of fundamental fairness120 or identifying the vested rights doctrine as balancing the power of government to control land use with the
substantive rights of property owners.121 This hodgepodge of justifications all support the underlying rule that once a right has vested it is
generally immune from government interference without compensation, but the confusion about the source of the rule makes the precise
content of the protection difficult to discern.122
119 See, e.g., Colo. Ground Water Comm’n v. North Kiowa-Bijou Groundwater Mgmt.
Dist., 77 P.3d 62, 81 (Colo. 2003) (finding that plaintiff’s groundwater right was vested right
and “[a]s such it cannot be taken away by subsequent legislative action”); First of Am.
Trust Co. v. Armstead, 664 N.E.2d 36, 40 (Ill. 1996) (“[T]his court has defined a vested
right as an expectation that is so far perfected that it cannot be taken away by legislation.”); Resolution Trust Corp. v. Fleischer, 892 P.2d 497, 501 (Kan. 1995) (“We, like many
courts, have used the term ‘vested rights’ to describe rights which cannot be taken away by
retroactive legislation.”); Powell v. Calvert County, 795 A.2d 96, 102 (Md. 2002) (“[Vested
rights] doctrine . . . rests upon the legal theory that when a property owner . . . completes
substantial construction on [his] property, his right to complete and use that structure
cannot be affected by any subsequent change of the applicable building or zoning regulations.” (quoting Prince George’s County v. Equitable Trust Co., 408 A.2d 737, 741 (Md. Ct.
Spec. App. (1979))).
120 See Daniel R. Mandelker, Entitlement to Substantive Due Process: Old Versus New
Property in Land Use Regulation, 3 WASH. U. J.L. & POL’Y 61, 76 n.71 (2000) (collecting
cases demonstrating that municipality’s revocation of building permit from property owner
with vested right in permit constitutes violation of substantive due process); see also, e.g.,
Goldrush II v. City of Marietta, 482 S.E.2d 347, 358 (Ga. 1997) (“A property interest protected by the due process clauses of the federal and state constitutions meets our definition
of ‘vested rights.’”); Hayes v. Howell, 308 S.E.2d 170, 175 (Ga. 1983) (noting that “vested
rights” is generally used to imply interests which “may not be interfered with by retrospective laws,” which “it is proper for the state to recognize and protect, and of which the
individual cannot be deprived arbitrarily without injustice.” (quoting Am. States Water
Serv. Co. of Cal. v. Johnson, 88 P.2d 770, 774 (Cal. Dist. Ct. App. 1939))); Overstreet &
Kirchheim, supra note 74, at 1071–72 (“Some cases couch the constitutional purpose of the
Washington rule in terms of how vesting provides citizens with ‘fundamental fairness,’
which is a due process concept.”).
121 See, e.g., Waikiki Marketplace Inv. Co. v. Honolulu, 949 P.2d 183, 193–94 (Haw. Ct.
App. 1997) (“[D]ue process principles protect a property owner from having his or her
vested property rights interfered with and preexisting lawful uses of property are generally
considered to be vested rights that zoning ordinances may not abrogate.” (citation
omitted)); Michael Weimann Assocs. Gen. P’ship v. Town of Huntersville, 555 S.E.2d 342,
345 (N.C. Ct. App. 2001) (“At common law, the vested rights doctrine evolved as a balancing mechanism” between “the State’s constitutional authority over land-use” and property owners’ “constitutional entitlement to due process of law”); Weyerhaeuser v. Pierce
County, 976 P.2d 1279, 1284 (Wash. Ct. App. 1999) (“The doctrine is based upon constitutional principles of fairness and due process, acknowledging that development rights are
valuable and protected property interests.”).
122 Kainen, supra note 117, at 120–22 (discussing vested rights in land use). According
to Kainen, whether a right has vested should not determine whether the right is subject to
1248
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
Finally, the constitutional inquiry at the heart of the nuisance
exception is different still. Some courts apply a relatively straightforward takings analysis, in essence finding that harm prevention provides a safe harbor from takings liability. These courts reason that the
government can regulate away a nuisance or harm because the owner
never had a property right to create the nuisance or harm in the first
place.123 Such a regulation is not a taking simply because it is not
taking a property right that the owner ever possessed. Other courts,
however, appear to rely on a substantive due process analysis. They
reason that harm prevention is a valid exercise of a state’s police
power and that, since all property is owned subject to the police
power, no such harm prevention can trigger a compensation requirement.124 Fundamentally, the inquiry in these latter opinions is
whether the regulation is an invalid exercise of the police power—that
is, whether it is irrational or arbitrary.
Underlying all of these courts’ opinions is a firm conviction that
regulations of existing uses raise constitutional issues. Which issues,
however, and how they arise, remains remarkably up for grabs. There
is no doubt that courts invoke both the Takings Clause and the Due
Process Clause to protect existing uses. The question, then, is the
extent to which these constitutional doctrines actually compel that
protection. Unpacking current takings and due process law only
reveals more confusion around the nature and extent of the
protection.
subsequent regulation. As he succinctly concludes, “Establishing that development expectations are ‘vested rights’ is neither a necessary nor a sufficient condition to their constitutional protection.” Id. at 122.
123 See, e.g., Lucas, 505 U.S. at 1029 (“[The limitation imposed by] regulations that prohibit all economically benficial use of land . . . must inhere in the title itself, in the restrictions that background principles of the State’s law of property and nuisance already place
upon land ownership.”).
124 See, e.g., City of Minot v. Freelander, 426 N.W.2d 556 (N.D. 1988) (holding that
demolition of house to stop nuisance does not violate Takings Clause); Just v. Marinette
County, 201 N.W.2d 761 (Wis. 1972) (upholding wetlands regulation as preventing harm).
But see Dep’t of Agric. & Consumer Servs. v. Polk, 568 So. 2d 35, 48 (Fla. 1990) (Barkett,
J., concurring) (“Although this Court has applied the harm-benefit distinction to determine liability, I now believe that analysis is inappropriate in ‘takings’ cases.” (citations
omitted)). The relationship between the police power and the Takings Clause is enormously contested. See Thomas W. Merrill, The Economics of Public Use, 72 CORNELL L.
REV. 61, 70 (1986) (discussing this relationship in context of Supreme Court efforts to
define Takings Clause’s ‘public use’ requirement); James E. Krier & Christopher Serkin,
Public Ruses, 2004 MICH. ST. L. REV. 859, 862–63 (2004) (discussing Merrill, supra, and
arguing that if public use were coterminous with police power it would not eliminate compensation inquiry).
November 2009]
EXISTING USES
1249
B. The Takings Clause
There are two principal bases for takings liability that are relevant to existing uses.125 The first is the per se rule from Lucas v. South
Carolina Coastal Council.126 The second, and the more important, is
the ad hoc balancing test from Penn Central.127 Both implicate the
protection of existing uses but neither provides for their categorical
protection.
The Lucas rule is narrow in scope and easy to state. According to
the Supreme Court, “Where the State seeks to sustain regulation that
deprives land of all economically beneficial use, we think it may resist
compensation only if . . . the proscribed use interests were not part of
[the owner’s] title to begin with.”128 Provocatively, the test focuses on
use rather than market value, implicating the protection of existing
uses.129 Of course, the Lucas test does not create per se takings liability whenever the government eliminates an existing use, but only
when it eliminates all economically beneficial uses. While the test
acknowledges the importance of use as opposed to value, it does not
provide any kind of comprehensive protection for existing uses.
Penn Central provides more protection for existing uses. In fact,
as originally formulated by the Supreme Court, the Penn Central test
appears to be principally concerned with existing uses, although subsequent iterations of the test have lost this focus. In Penn Central, the
Supreme Court identified three factors for evaluating whether a regulation amounts to a taking: (1) the extent of the diminution in property value it caused, (2) its interference with reasonable investmentbacked expectations, and (3) its character.130 The first and second
factors are at least potentially relevant to the protection of existing
uses and are most usefully considered in reverse order.131
125 This assumes that existing uses are not eliminated through permanent physical occupations, proscribed in Loretto v. Teleprompter Manhattan CATV Corp., 458 U.S. 419, 421
(1982), nor eliminated through government exactions, the permissible limits of which are
defined in Dolan v. City of Tigard, 512 U.S. 374 (1994), and Nollan v. Cal. Coastal
Comm’n, 483 U.S. 825 (1987).
126 505 U.S. 1003, 1015 (1992) (deeming “categorical treatment appropriate . . . where
regulation denies all economically beneficial or productive use of land”).
127 Penn Cent. Transp. Co. v. City of New York, 438 U.S. 104 (1978).
128 Lucas, 505 U.S. at 1027.
129 See STEVEN J. EAGLE, REGULATORY TAKINGS §7-3(b)(5) (3d ed. 2005) (discussing
Lucas Court’s emphasis on economically beneficial use of property).
130 Penn Central, 438 U.S. at 124.
131 The third factor might also be relevant to this analysis, at least as interpreted by
Professor Eagle. Traditionally, the character of the regulation implicated only a distinction
between a physical invasion and regulation merely affecting economic interests. Steven J.
Eagle, “Character” as “Worthiness”: A New Meaning for Penn Central’s Third Test?, 27
ZONING & PLANNING L. REP. 1, 2 (2004). Professor Eagle notes, however, that more
1250
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
The diminution in value test has become the focus of most takings analysis, but it may be Penn Central’s second factor—the interference with investment-backed expectations—that provides the most
significant protection for existing uses.132 The facts of the Penn
Central case are well known; nevertheless, a quick reminder is important to understand the original application of the second factor.
In 1967, New York City designated Grand Central Terminal a
landmark and therefore subject to its Landmarks Preservation Law.133
This designation prevented Penn Central, the owner of Grand
Central, from developing an office tower on top of the terminal.134
Rejecting Penn Central’s regulatory takings claim, the Court focused
on the fact that the landmark designation did not interfere with Penn
Central’s existing use of the building but only with prospective plans.
As the Court held:
[T]he New York City law does not interfere in any way with the
present uses of the Terminal. Its designation as a landmark not only
permits but contemplates that appellants may continue to use the
property precisely as it has been used for the past 65 years: as a
railroad terminal containing office space and concessions. So the
law does not interfere with what must be regarded as Penn Central’s
primary expectation concerning the use of the parcel.135
In fact, the Court expressly distinguished other cases on grounds that
they involved an interference with the present existing use of property.136 According to the Supreme Court, then, the existing use of
property constitutes an owner’s “primary expectation” about the use
of that property.137
recently, some courts have expanded the inquiry to include the purpose and benefits of the
government action, specifically looking at whether the government regulation is retroactive or targeting an individual. Id. at 4–5. This reading of the third Penn Central factor is
very interesting but does not represent a consensus view of the content of the test.
132 See Mark W. Cordes, Takings Jurisprudence as Three-Tiered Review, 20 J. NAT.
RESOURCES & ENVTL. L. 1, 38 (2005) (“[T]he only clear case for a taking under Penn
Central would be where the investment reflects actual development expenditures, such as
constructing facilities or homes, rather than speculation on future uses.”); Michael B. Kent
Jr., Construing the Canon: An Exegesis of Regulatory Takings Jurisprudence After Lingle v.
Chevron, 16 N.Y.U. ENVTL. L.J. 63, 97–98 (2008) (identifying “subjective” aspects of
investment-backed expectations, such as extent and characteristics of property owner’s
actual investment, and noting similarity to vested rights inquiry).
133 Penn Central, 438 U.S. at 115–16.
134 Id. at 116–18.
135 Id. at 136.
136 See id. (distinguishing Goldblatt v. Hempstead, 369 U.S. 590 (1962), Miller v.
Schoene, 276 U.S. 272 (1928), United States v. Causby, 328 U.S. 256 (1946), Griggs v.
Allegheny County, 369 U.S. 84 (1962), and Hadacheck v. Sebastian, 239 U.S. 394 (1915)).
137 See id. The Court’s presumption about existing uses, however, may not be accurate.
Existing uses can be speculative uses, and despite their lack of immediate utility or function they may simply be creative methods of land speculation. Parking lots in urban cen-
November 2009]
EXISTING USES
1251
Over time, the inquiry into investment-backed expectations has
changed and lost its focus on the existing use of the property. In
Kaiser Aetna v. United States,138 the Supreme Court subtly but profoundly changed the Penn Central language from “distinct investmentbacked expectations”139 to “reasonable investment backed expectations.”140 Partly as a result, this factor is now principally used to
distinguish a property owner’s reasonable expectations from pie-inthe-sky development dreams.141 There is no taking, for example, if a
developer buys a small lot in a residential area and is prohibited from
building a large strip mall or gas station, because that expectation was
unreasonable.142 The Takings Clause will not protect unreasonable
expectations about the use of property, even if prohibiting that use
results in a significant diminution in value.143 In other words, the test
today focuses more on the reasonableness of a property owner’s
ters are one such example. See Jane Jacobs, Remarks, Random Comments, 28 B.C. ENVTL.
AFF. L. REV. 537, 541 (2001) (“Land in potentially lucrative and productive locations that
is being kept in low-value parking lots is a symptom of land speculation: land remaining
almost idle in hope that its very emptiness will increase its value to some future buyer.”).
138 444 U.S. 164 (1979).
139 Penn Central, 438 U.S. at 124 (emphasis added).
140 Aetna, 444 U.S. at 175 (emphasis added). For further analysis of this transformation,
see Steven J. Eagle, The Regulatory Takings Notice Rule, 24 U. HAW. L. REV. 533, 560
(2002) (characterizing transformation as move from subjective to objective standard).
141 See J. David Breemer, Playing the Expectations Game: When Are Investment-Backed
Land Use Expectations (Un)reasonable in State Courts?, 38 URB. LAW. 81, 85 (2006)
(“Together, these decisions redirected the expectations inquiry away from the impact of
regulation and toward the appropriateness of the landowners’ land use expectations . . . .”);
Mandelker, supra note 36, at 21 (“[T]he investment-backed expectations factor has
become . . . a shield for government that protects land use regulations from the Takings
Clause.”). The term “expectations” suffers from some of the same problems as “vested
rights.” “[C]ourts often use ‘expectation’ to refer to an interest less deserving of protection
than a ‘right.’ . . . By the same token, however, the Court has also used the word ‘expectations’ to refer to protected property interests.” Lynda J. Oswald, Cornering the Quark:
Investment-Backed Expectations and Economically Viable Uses in Takings Analysis, 70
WASH. L. REV. 91, 108 (1995).
142 Cf. Oswald, supra note 141, at 111–12 (discussing property owner’s expectations as
bound up with ability to anticipate future government regulation).
143 See, e.g., Nat’l Viatical, Inc. v. Oxendine, No. 1:05-CV-3059-TWT, 2006 WL 1071839,
at *4 (N.D. Ga. Apr. 20, 2006) (dismissing complaints based on allegations that statute
lowered prices of life insurance policies on secondary market because plaintiff failed “to
demonstrate how this regulation interferes with their reasonable investment-backed expectations”); Peoples Super Liquor Stores, Inc. v. Jenkins, 432 F. Supp. 2d 200, 215–16 (D.
Mass. 2006) (holding that prohibition against owning interest in more than three liquor
stores did not interfere with franchisee’s reasonable investment-backed expectations); Bd.
of Zoning Appeals v. Leisz, 702 N.E.2d 1026, 1030 (Ind. 1998) (noting that “the Supreme
Court has ‘uniformly reject[ed] the proposition that diminution in property value, standing
alone, can establish a taking’” and opining that “[i]n particular, the forfeiture of [plaintiff’s] nonconforming use caused no interference with their reasonable investment-based
expectations” (internal citation omitted)).
1252
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
expectations, not on her investments, and therefore focuses on future
uses and not existing ones.144
Nowhere is the confusion about the meaning of investmentbacked expectations more immediately apparent than in the Supreme
Court’s decision in Palazzolo v. Rhode Island.145 There, the Court
wrestled with the question whether the transfer of property after
enactment of a regulation eliminates any takings claim because the
regulation becomes part of the new owner’s reasonable expectations.146 The majority held that takings claims survive postenactment
transfer of the property.147 In concurrence, Justice O’Connor argued
that the inquiry into investment-backed expectations is ad hoc and
multifactored and that preexisting regulations are one factor among
many for courts to consider.148 In contrast, Justice Scalia concurred
separately to write that preexisting regulations should have no impact
on a property owner’s investment-backed expectations.149 Justice
Stevens, dissenting, argued exactly the opposite: that preexisting regulations should be dispositive as to the takings claim.150 The content
and significance of reasonable investment-backed expectations are not
easily defined.
The application of investment-backed expectations specifically to
existing uses is similarly up for grabs. Some have suggested that it
should be used to limit protection for property acquired by gift or
inheritance because of the absence of investment, an interpretation
completely removed from existing use protection.151 Professor
Mandelker has argued that investment-backed expectations might be
144 E.g., 26 AM. JUR. 2D Eminent Domain § 14 (2004) (“[I]nvestment-backed expectation must be reasonable.”).
145 533 U.S. 606 (2001).
146 See id. at 626 (characterizing State’s position as argument that “by prospective legislation the State can shape and define property rights and reasonable investment-backed
expectations, and subsequent owners cannot claim any injury from lost value” because
they “purchased or took title with notice of the limitation”).
147 Id. at 629–30.
148 See id. at 634–36 (O’Connor, J., concurring) (“[T]he state of regulatory affairs at the
time of acquisition is not the only factor that may determine the extent of investmentbacked expectations. . . . Courts properly consider the effect of existing regulations under
the rubric of investment-backed expectations in determining whether a compensable
taking has occurred. As before, the salience of these facts cannot be reduced to any ‘set
formula.’”).
149 See id. at 637 (Scalia, J., concurring) (“[T]he fact that a restriction existed at the time
the purchaser took title . . . should have no bearing upon the determination of whether the
restriction is so substantial as to constitute a taking.”).
150 See id. at 641 (Stevens, J., dissenting in part) (“If the regulations imposed a compensable injury on anyone, it was on the owner of the property at the moment the regulations
were adopted.”).
151 See, e.g., Richard A. Epstein, Lucas v. South Carolina Coastal Council: A Tangled
Web of Expectations, 45 STAN. L. REV. 1369, 1370 (1993) (suggesting “investment-backed
November 2009]
EXISTING USES
1253
entitled to protection only against substantial and unforeseeable regulations.152 Professor Epstein has argued that “investment-backed
expectations” is an unfortunate turn of phrase, actually synonymous
with “reasonable expectations” or even with “private property.”153
Some commentators have suggested that it is not particularly different
from diminution in value—Penn Central’s first factor.154 Courts have
fared no better at providing coherent content to this test.155 All this is
to say that Penn Central’s focus on investment-backed expectations
may, at least under some interpretations, protect existing uses some of
the time. However, because the content and application of this factor
is ad hoc at best and hopelessly confused at worst, its actual application to existing uses is less than conclusive.156
The first Penn Central factor, diminution in value, provides even
less doctrinal justification for protecting existing uses. This test has,
over time, become the cornerstone of takings jurisprudence.157
Applying this factor, a regulation is a compensable taking if it reduces
expectations” may be term excluding property received by gift); Oswald, supra note 141, at
116–17 (discussing Hodel v. Irving, 481 U.S. 704 (1987)).
152 See Mandelker, supra note 36, at 25.
153 See Epstein, supra note 151, at 1370 (“All in all, we should be deeply suspicious of
the phrase ‘investment-backed expectations’ because it is not possible to identify even the
paradigmatic case of its use.”).
154 See, e.g., Oswald, supra note 141, at 101–06 (examining history of investment-backed
expectations). It appears that Professor Michelman was the first person to use the phrase,
“investment-backed expectations” in the takings context. Frank I. Michelman, Property,
Utility, and Fairness: Comments on the Ethical Foundations of “Just Compensation” Law,
80 HARV. L. REV. 1165, 1213 (1967). He discussed the concept in the context of examining
diminution in value. Id. at 1229–34.
155 See Breemer, supra note 141, at 82 (“Trying to determine when land use expectations
are reasonable in state courts is an experience akin to a dog chasing its own tail.”); see also,
e.g., Dana, supra note 40, at 661 (“[C]ourts provide very little explanation for their holdings as to when it is and is not reasonable for a property owner to expect that she will be
subject to uncompensated regulation in the future.”).
156 See Epstein, supra note 151, at 1370 (“All in all, we should be deeply suspicious of
the phrase ‘investment-backed expectations’ because it is not possible to identify even the
paradigmatic case of its use.”); Oswald, supra note 141, at 107 (“[T]he meaning of the
phrase remains uncertain, rendering its effectiveness as a legal doctrine questionable at
best.”); see also Breemer, supra note 141, at 81–82 & n.6 (citing sources and describing
investment-backed expectations as “certainly the least understood” Penn Central factor).
157 See Oswald, supra note 141, at 130 (“Over the past sixteen years, the Supreme Court
has increasingly committed itself to a regulatory takings analysis that focuses extensively, if
not exclusively, upon the economic effects of the regulation upon the property owner.”);
Joseph L. Sax, Takings, Private Property, and Public Rights, 81 YALE L.J. 149, 151 (1971)
(“[T]he criterion for recognizing a particular economic injury which follows from government action as a taking is the extent of economic loss.”). But see Mayhew v. Town of
Sunnyvale, 964 S.W.2d 922, 935 (Tex. 1998) (identifying focus of Penn Central as “the
extent to which the regulation has interfered with distinct investment-backed expectations”); Eagle, supra note 140, at 561 (“[A]t least in the minds of some courts, ‘investmentbacked expectations’ would crowd out the other prongs of the three-factor test.”).
1254
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
the value of the property by too much.158 Under this test, a regulation
of an existing use is more likely to be a compensable taking than a
regulation of a prospective future use. Imagine two parcels of land
identical but for the fact that the first has a building on it and the
second does not. Imagine further that the government passes a new
regulation that requires eliminating the building on the first parcel
and prevents a building from ever being built on the second. The
owner of the first parcel has suffered a more significant economic
impact—the regulation reduced the value of his land more than it
reduced the value of his neighbor’s land.159
Diminution in value therefore goes some way towards protecting
existing uses, but it is wholly inadequate to support the kind of comprehensive protection that property law exhibits. Depending on the
residual value remaining in the regulated property, many existing uses
could still be eliminated without triggering liability. Because the Penn
Central test is applied in reference to the value of the property as a
whole,160 it serves only to prevent the government from eliminating
valuable existing uses on parcels of land that have little other value.
Where the existing use either does not add much to the financial value
of the underlying property—imagine run-down, low-income housing
in a gentrifying neighborhood—or where an existing use is but one
part of a much larger property, the diminution in value test provides
little protection.
Of course, eliminating an existing use would always trigger liability under the diminution in value test if the property at issue is
defined as the use itself. This presents the familiar conceptual severance problem about how to define the relevant property.161 Professor
158 What counts as too much has been a source of constant controversy, but the percentage diminution required appears to be quite high. Cordes, supra note 132, at 39 (surveying cases and finding that “diminution in value must substantially exceed 50%, and
should be closer to 90%, before any serious consideration is given of a . . . taking”).
159 Before the regulation, the first lot might have been worth $1,000,000 and the second
lot worth $250,000 (reflecting the absence of a building). After the regulation, they are
both worth the same amount, say $50,000 (excepting the transition costs of removing the
existing building, which this stylized hypothetical excludes, but which are considered in
detail below). The diminution in value of the first lot is 95% and of the second is only
80%. Perversely, however, a regulation that has a fixed economic impact is more likely to
be a taking when applied to undeveloped property because the denominator will be
smaller; a regulation imposing a $50,000 loss may be a taking of property worth $60,000,
but is unlikely to be a taking of property worth $200,000.
160 See Penn Cent. Transp. Co. v. City of New York, 438 U.S. 104, 130–31 (1978) (“In
deciding whether a particular governmental action has effected a taking, this Court focuses
[on] . . . the parcel as a whole . . . .”).
161 For a discussion of conceptual severance, see Margaret Jane Radin, The Liberal Conception of Property: Cross Currents in the Jurisprudence of Takings, 88 COLUM. L. REV.
1667, 1676 (1988) (“This strategy I shall call ‘conceptual severance.’ It consists of
November 2009]
EXISTING USES
1255
Michelman, whose incomparably influential article on the Takings
Clause provided a blueprint for the Penn Central test, observed precisely this point in anticipating the special protection for existing uses:
What explains, then, the universal understanding that only those
nonconforming uses are protected which were demonstrably afoot
by the time the regulation was adopted? The answer seems to be
that actual establishment of the use demonstrates that the prospect
of continuing it is a discrete twig out of his fee simple bundle to
which the owner makes explicit reference in his own thinking, so
that enforcement of the restriction would, as he looks at the matter,
totally defeat a distinctly crystallized expectation.162
By singling out the use as a protectable stick in the bundle of property
rights, Michelman seems to have been responding to this same central
intuition that existing uses are entitled to property protection. Nevertheless, most courts take a much broader view when defining the
property at issue.163 The Supreme Court itself has held that diminution in value is to be applied to the parcel as a whole.164 While this
delineating a property interest consisting of just what the government action has removed
from the owner, and then asserting that that particular whole thing has been permanently
taken.”); see also Vicki Been & Joel C. Beauvais, The Global Fifth Amendment? NAFTA’s
Investment Protections and the Misguided Quest for an International “Regulatory Takings”
Doctrine, 78 N.Y.U. L. REV. 30, 64 (2003) (discussing Radin, supra).
162 Michelman, supra note 154, at 1233. Michelman’s article, usually cited for his proposed utilitarian formulation of takings liability, remains the single most influential article
on the Takings Clause. A number of people have observed that it provided the blueprint
for Penn Central. See Oswald, supra note 141, at 104 (“Michelman’s analysis clearly influenced Justice Brennan as he wrote the majority opinion in Penn Central.”); Radin, supra
note 161, at 1684 (describing Penn Central test as “created under the salutary influence of
Frank Michelman’s famous article”); Gregory M. Stein, The Effect of Palazzolo v. Rhode
Island on the Role of Reasonable Investment-Backed Expectations, in TAKING SIDES ON
TAKINGS ISSUES: PUBLIC AND PRIVATE PERSPECTIVES, supra note 36, at 41–42 (“The
Court relied heavily on Professor Frank Michelman’s influential 1967 essay . . . .”).
163 See Giovanella v. Conservation Comm’n of Ashland, 857 N.E.2d 451, 456 (Mass.
2006) (“When a court considers a large piece of land of which only a small portion has lost
value due to regulation, it is less likely to conclude that a taking has occurred. If a court
considers a smaller parcel of land, most of which has been affected by a regulation, then
the economic impact is more likely to appear large enough to constitute a taking.”); see
also City of Coeur d’Alene v. Simpson, 136 P.3d 310, 319 (Idaho 2006) (“Courts typically
reject the so-called ‘conceptual severance’ theory—the notion that whole units of property
may be divided for the purpose of a takings claim.”); Smith v. Town of Mendon, 822
N.E.2d 1214, 1220 n.12 (N.Y. 2004) (evaluating “effect of the government action on the
value of the property as a whole, rather than . . . its effect on discrete segments of the
property”); Machipongo Land & Coal Co. v. Commonwealth, 799 A.2d 751, 768 (Pa. 2002)
(applying Penn Central’s mandate to consider the “parcel as a whole”).
164 See Penn Central, 438 U.S. at 131 (focusing on “parcel as a whole”); see also TahoeSierra Pres. Council, Inc. v. Tahoe Reg’l Planning Agency, 535 U.S. 302, 331 (2002)
(rejecting temporal division of property); Andrus v. Allard, 444 U.S. 51, 65–66 (1979) (“At
least where an owner possesses a full ‘bundle’ of property rights, the destruction of one
‘strand’ of the bundle is not a taking, because the aggregate must be viewed in its
entirety.”).
1256
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
definition has defied efforts at precise line drawing, it nevertheless
severely limits the protection that diminution in value affords to
existing uses.
It is important to point out, too, that the Court decided Penn
Central in 1978. The various land use doctrines discussed in Part I
predate Penn Central by as much as 50 years. To the extent they
assume that existing uses are generally immune from government regulation, this assumption does not come from Penn Central, nor, even,
from the Takings Clause, which had very limited scope throughout the
early part of the twentieth century.165 Instead, the same background
intuition about existing uses animates both land use doctrine and the
Takings Clause while its original source remains unclear.
C. The Due Process Clause
The Due Process Clause is the other principal source of constitutional protection for private property. Substantive due process, in
particular, includes some degree of protection for existing uses,
although the story here is quite complex.166 After the fallout from
165 See Matthew P. Harrington, “Public Use” and the Original Understanding of the SoCalled “Takings” Clause, 53 HASTINGS L.J. 1245, 1247 (2002) (“[T]he idea that courts had
the power to supervise legislative expropriations would have been unfamiliar to the members of the Congress who drafted the so-called Takings Clause.”); see also Lucas v. S.C.
Coastal Council, 505 U.S. 1003, 1057 n.23 (1992) (“James Madison, author of the Takings
Clause, apparently intended it to apply only to direct, physical takings of property by the
Federal Government” (citing William Michael Treanor, Note, The Origins and Original
Significance of the Just Compensation Clause of the Fifth Amendment, 94 YALE L.J. 694,
711 (1985))). But see Nicole Stelle Garnett, “No Taking Without a Touching?” Questions
from an Armchair Originalist, 45 SAN DIEGO L. REV. 761, 762–64 (2008) (questioning historical accounts of Takings Clause).
166 See Stewart E. Sterk, The Federalist Dimension of Regulatory Takings Jurisprudence,
114 YALE L.J. 203, 267 n.292 (2004) (“In addition to the Takings Clause, the [New York
Court of Appeals] has also invoked substantive due process to protect landowners from
actions by local zoning authorities.”); Mark Tunick, Constitutional Protections of Private
Property: Decoupling the Takings and Due Process Clauses, 3 U. PA. J. CONST. L. 885, 899
(2001) (“The Due Process Clauses provide protections against unfair laws, both those that
take property as well as those that merely regulate it. The Due Process Clause, not the
Takings Clause, protects citizens from state laws that are overbroad, arbitrary, insufficiently justified, or unfair.”); Robert Ashbrook, Comment, Land Development, the
Graham Doctrine, and the Extinction of Economic Substantive Due Process, 150 U. PA. L.
REV. 1255, 1257 (2002) (noting in land use context alone “federal courts have allowed
economic substantive due process . . . to escape extinction (and in some instances even to
flourish)”); see also Jane B. Baron, Winding Toward the Heart of the Takings Muddle:
Kelo, Lingle, and Public Discourse About Private Property, 34 FORDHAM URB. L.J. 613,
640 (2007) (describing post-Lingle “sequencing between the analyses under the two
clauses, with substantive due process questions to be raised first and, assuming the statute
in question withstands due process scrutiny (a safe assumption for economic regulation),
takings scrutiny to follow”). See generally Mandelker, supra note 120 (arguing for more
frequent use of substantive due process analysis because of inadequacy of takings law).
November 2009]
EXISTING USES
1257
Lochner v. New York,167 courts have had little appetite for striking
down economic regulations on substantive due process grounds.168
After that case was overruled, substantive due process protection for
most economic rights all but ended.169 Curiously, land use is an area
where courts have consistently remained willing to invalidate regulations under substantive due process.170
As the Supreme Court recently explained, the Due Process
Clause prevents government regulations that are arbitrary or irrational, including regulations that have an insufficient means-ends
fit.171 This rule, in turn, requires courts to judge the government’s
action against its goals.172 Due process analysis will therefore prohibit
the government from imposing harms that are disproportionately high
compared to the benefits created. Considerable confusion remains.173
Due process review is highly deferential, and land use controls
enacted pursuant to the government’s police power are entitled to a
167
198 U.S. 45 (1905).
See Richard Squire, Antitrust and the Supremacy Clause, 59 STAN. L. REV. 77, 104
(2006) (“Judges across the ideological spectrum now try to outdo each other in denouncing
Lochner, marking a broad consensus that principles whereby courts use federal law to cut
down state economic regulation should not originate in the judicial branch.”); Tunick,
supra note 166, at 899 n.59 (“Since the demise of Lochner, the Court has been reluctant to
strike down economic legislation on due process grounds . . . .”).
169 See, e.g., Ferguson v. Skrupa, 372 U.S. 726, 729–31 (1963) (“It is now settled that
States have power to legislate against what are found to be injurious practices in their
internal commercial and business affairs, so long as their laws do not run afoul of some
specific federal constitutional prohibition, or of some valid federal law.” (internal quotation marks and citations omitted)).
170 See, e.g., Marks v. City of Chesapeake, 883 F.2d 308, 313 (4th Cir. 1989) (holding that
withholding permit for palm reading business is arbitrary and capricious and violates due
process); Sullivan v. Town of Salem, 805 F.2d 81, 85 (2d Cir. 1986) (holding that prohibiting
occupancy of new home that meets building code violates due process); see also Michael J.
Phillips, The Slow Return of Economic Substantive Due Process, 49 SYRACUSE L. REV. 917,
927–36 (1999) (citing land use decisions); Anthony B. Sanders, The New “Judicial Federalism” Before Its Time: A Comprehensive Review of Economic Substantive Due Process
Under State Constitutional Law Since 1940 and the Reasons for Its Recent Decline, 55 AM.
U. L. REV. 457, 469 (2005) (“Instead of a deferential rational basis test, where the regulation at issue is heavily presumed to be constitutional, courts often apply a mere ‘clear and
convincing evidence’ presumption when deciding land use zoning cases.”).
171 Lingle v. Chevron, 544 U.S. 528, 542 (2005) (“[A] regulation that fails to serve any
legitimate governmental objective may be so arbitrary or irrational that it runs afoul of the
Due Process Clause.”). Other courts have made similar statements. See, e.g., DeBlasio v.
Zoning Bd. of Adjustment, 53 F.3d 592, 593 (3d Cir. 1995) (“[I]n the context of land use
regulation, a property owner states a substantive due process claim where he or she alleges
that the decision limiting the intended land use was arbitrarily or irrationally reached.”).
For further discussion of Lingle, see infra notes 192–94 and accompanying text.
172 See, e.g., Lingle, 544 U.S. at 541 (describing means-ends test).
173 See, e.g., Ashbrook, supra note 166, at 1260–61 (“The precise legal standard varies
by circuit. The District of Columbia Circuit merely requires a showing of ‘grave unfairness.’ More commonly, however, courts require some variation on arbitrariness and/or
capriciousness.” (footnotes omitted)).
168
1258
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
presumption of validity.174 Nevertheless, some courts have found land
use regulations irrational simply by dint of their impact on existing
uses.175
More typically, however, the due process inquiry into means-ends
fit is only tangentially related to existing uses. Indeed, it is easy to
imagine many situations in which a municipality’s decision to eliminate an existing use is both perfectly rational and likely to succeed in
achieving its goals. Consider, for example, a town that wants to apply
its zoning ordinance to prior nonconforming uses, say to eliminate
existing businesses from a neigborhood newly zoned residential. The
goal of separating uses is decidedly rational, and imposing that separation on existing uses will be even more effective at achieving this
goal.176
Therefore, the due process inquiry itself does not appear to provide particularly strong protection for existing uses. Its balancing of
benefits and burdens is effective at proscribing arbitrarily short amortization periods or arbitrary distinctions in the development process
that determine whether rights vest. But it does not explain why amortization periods are required at all or why vested rights are entitled to
protection against legislative change.
D. The Relationship Between Takings and Due Process
Undoubtedly, part of the problem of identifying any constitutional basis for the protection of existing uses is the historically complicated relationship between the Takings Clause and the Due Process
Clause.177 Courts and commentators traditionally distinguished
between valid exercises of the police power, on the one hand, and
impermissible takings of private property, on the other.178 Under this
174 See, e.g., Amnon Lehavi, Intergovernmental Liability Rules, 92 VA. L. REV. 929, 937
(2006) (“Courts deem [land use regulations] to be presumptively valid, and place the
burden of proof on those making sustantive due process claims to show that the regulation
is arbitrary and unreasonable.”).
175 See, e.g., Eger v. Levine, 545 N.Y.S.2d 618, 619 (App. Div. 1989) (finding denial of
use variance to be irrational where use predated effective date of zoning ordinance). For
an example of a court rejecting a due process challenge to a regulation eliminating an
existing use, see Hartland Sportsman’s Club, Inc. v. Town of Delafield, 35 F.3d 1198 (7th
Cir. 1994).
176 See supra note 8 and accompanying text (noting that original justification of zoning
was to separate incompatible uses of property).
177 For a fascinating and thoroughgoing analysis of this history, see Karkkainen, supra
note 100, at 838–51.
178 See, e.g., Lincoln Trust Co. v. Williams Bldg. Corp., 128 N.E. 209, 210 (N.Y. 1920)
(“In a great metropolis like New York, in which the public health, welfare, convenience,
and common good are to be considered, I am of the opinion that the resolution was not an
incumbrance, since, it was a proper exercise of the police power. The exercise of such
power, within constitutional limitations, depends largely upon the discretion and good
November 2009]
EXISTING USES
1259
view, property is owned subject to the state’s police power, which
includes the power to redefine the content of property rights.179 The
principal constitutional inquiry is therefore focused on the validity of
the state’s exercise of its police power and not on the impact of a
regulation on property. In other words, courts evaluating the constitutionality of state land use laws or regulations need only decide
whether the government acted within the scope of its permissible
police powers, and if so, there is no taking.
As a result, cases from the early part of the twentieth century
interpreting the extent of state and local governments’ power to zone
focused almost exclusively on the justification for the regulation and
the means the government employed to accomplish those ends.180
Courts would invalidate regulations that they found were arbitrary or
irrational.181 Often, this involved a court’s judgment about whether or
not the regulation was somehow unfair.182 Fundamentally, however,
this inquiry was rooted in the Due Process Clause and not the Takings
Clause.183
Despite some foreshadowing in 1922,184 the Takings Clause did
not come into its own until 1978 in Penn Central.185 The Penn Central
judgment of the municipal authorities . . . .”); see also Merrill, supra note 124, at 69–70
(describing relationship between police power and public use); cf. Bettman, supra note 49,
at 835 (“The constitutional limitations upon eminent domain, such as the requirement that
compensation be paid, have no relevance where an ordinance is an exercise of the police
power.”).
179 See Karkkainen, supra note 100, at 841 (“Thus a legitimate exercise of the police
power could never give rise to a compensable taking . . . .” (footnote omitted)).
180 See, e.g., In re Opinion of the Justices, 127 N.E. 525, 527 (Mass. 1920) (“An ordinance or by-law which segregates manufacturing and commercial buildings on the one side,
from homes and residences on the other, is justified by the broad conceptions of the police
power . . . .”).
181 See, e.g., City of Tucson v. Ariz. Mortuary, 272 P. 923, 927 (Ariz. 1928) (referring to
Euclid and noting that ordinance is unconstitutional if it is “clearly arbitrary and unreasonable, and has not any substantial relation to the public health, safety, morals, or general
welfare”); Longley v. Rumsey, 224 N.Y.S. 165, 167 (Sup. Ct. 1927) (holding zoning ordinance invalid because of its failure to act in accordance with comprehensive plan and
“because of the arbitrary method by which the character of the so-called districts is determined”); Luse v. City of Dallas, 131 S.W.2d 1079, 1084 (Tex. Civ. App. 1939) (holding that,
to be unconstitutional and therefore void, “[an] ordinance . . . must itself be clearly arbitrary, unreasonable and without any substantial relation to the public health, safety,
morals, or general welfare”); Geisenfeld v. Vill. of Shorewood, 287 N.W. 683, 686 (Wis.
1939) (declaring ordinance “unconstitutional and void because ‘clearly arbitrary and
unreasonable, having no substantial relation to the public health, safety, morals, or general
welfare’” (quoting Vill. of Euclid v. Ambler Realty Co., 272 U.S. 365, 395 (1926))).
182 See Bettman, supra note 49, at 836 (“[I]n actual practice in constitutional cases, ‘reasonable’ often signifies little more than that . . . the balance of considerations of private
and public interests has been fairly maintained.”).
183 See Karkkainen, supra note 100, at 841–42 (defining content of due process review).
184 See supra note 99 and accompanying text.
185 Penn Cent. Transp. Co. v. City of New York, 438 U.S. 104 (1978).
1260
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
test, as described above, focuses entirely on the impact of the regulation on property rights.186 Two years later, however, in 1980, the due
process and takings tests came together in the Supreme Court’s decision in Agins v. City of Tiburon.187 In Agins, the Court held that
“[t]he application of a general zoning law to particular property
effects a taking if the ordinance does not substantially advance legitimate state interests, or denies an owner economically viable use of his
land.”188 As many commentators noted, this language appeared to
incorporate a substantive due process means-ends test into the
Takings Clause.189 For twenty-five years, courts wrestled with the
relationship between substantive due process and takings, often
applying a more searching inquiry under the latter than traditional
due process analysis would have required.190 They can therefore be
excused for invoking both the Takings Clause and substantive due
process in the same breath.191
Finally, in 2005, the Supreme Court backpedaled, overruling
Agins and holding in Lingle v. Chevron U.S.A. Inc. that the meansends test was exclusively a due process test.192 Lingle clarified the
focus of the takings and due process tests. The Due Process Clause
186 The three-part ad hoc balancing test focuses on the diminution in value, the extent of
interference with investment-backed expectations, and the character of the regulation. Id.
at 124.
187 447 U.S. 255 (1980), overruled in part by Lingle v. Chevron U.S.A. Inc., 544 U.S. 528,
540 (2005).
188 Id. at 260 (citation omitted).
189 See Karkkainen, supra note 100, at 883 (arguing that “[c]onflation of substantive due
process and Takings Clause doctrine muddled the ‘takings’ issue in multiple ways” and
pointing to Agins as point of departure from proper takings jurisprudence); see also David
L. Callies & Christopher T. Goodin, The Status of Nollan v. California Coastal
Commission and Dolan v. City of Tigard After Lingle v. Chevron, 40 J. MARSHALL L. REV.
539, 557 (2007) (“[T]he Agins test is essentially a substantive due process inquiry . . . .”);
Robert G. Dreher, Lingle’s Legacy: Untangling Substantive Due Process from Takings
Doctrine, 30 HARV. ENVTL. L. REV. 371, 372 (2006) (noting Supreme Court’s reliance, in
Agins, on “seminal due process cases”); Fenster, supra note 84, at 735 (“[T]he Agins ‘substantial advancement’ test . . . mistook a substantive due process test for a takings test
. . . .”).
190 See Nestor M. Davidson, The Problem of Equality in Takings, 102 NW. U. L. REV. 1,
16 (2008) (suggesting that Agins “substantially advances” language “formed the rhetorical
basis for the suggestion in Nollan that Takings Clause-based review of the impact of regulation on property rights should be undertaken through some form of heightened scrutiny”); Karkkainen, supra note 100, at 828 (noting “errant language in Agins that imported
Lochner-style heightened substantive due process review into modern takings law”).
191 For a very good history of the relationship between takings law and substantive due
process, see Dreher, supra note 189, at 373–87.
192 544 U.S. 528 (2005). For more in-depth discussion of the line of cases resulting in
Lingle, see generally Fenster, supra note 84. For a discussion of the doctrinal confusion
that existed prior to Lingle, see generally, for example, Thomas E. Roberts et al., LandUse Litigation: Doctrinal Confusion Under the Fifth and Fourteenth Amendments, 28 URB.
LAW. 765 (1996).
November 2009]
EXISTING USES
1261
prevents government regulations that are arbitrary or irrational,
including regulations that have an insufficient means-ends fit.193 In
contrast, the takings inquiry, according to the Court, is concerned only
with the extent of the government’s interference with property
rights.194
This history reveals that any doctrinal protection for existing uses
is built on shifting constitutional sands. Protection for existing uses
that may have been justified originally under substantive due process
now sounds more appropriately in the Takings Clause, which, as a
whole, entails inquiries that are less deferential to local government.
But neither today provides the kind of protection that land use law
implicitly assumes. Ultimately, current constitutional doctrine does
not compel categorical protection for existing uses.
NORMATIVE ACCOUNTS
III
EXISTING USE PROTECTION
OF
If the Constitution does not compel protection of existing uses,
do (or should) normative considerations justify the protection?
Courts—and people more generally—share a strong intuition that
existing uses should be protected. In the land use context, where real
property and preexisting investments in land are at stake, existing uses
seem to demand greater protection. Articulating precisely why, however, is harder than it might seem. In fact, none of the possible justifications for the categorical protection of existing uses withstands
serious scrutiny. The bulk of this Part is devoted to identifying, examining, and ultimately rejecting likely normative justifications for
existing use protection.
To be clear, many of the possible justifications for existing use
protection could apply in specific cases or contexts. The point here is
not to suggest that existing uses should never be protected—in fact,
they often should. The point, instead, is to demonstrate that they do
not deserve the kind of categorical constitutional protection courts
currently provide. Once existing uses are dismantled as a specially
protected category, it is possible to address with more precision,
nuance, and clarity whether a particular existing use should be
protected.
193 See Lingle, 544 U.S. at 542 (“[A] regulation that fails to serve any legitimate governmental objective may be so arbitrary or irrational that it runs afoul of the Due Process
Clause.”).
194 See id. at 542 (“In stark contrast to the three regulatory takings tests discussed
above, the ‘substantially advances’ inquiry reveals nothing about the magnitude or character of the burden a particular regulation imposes upon private property rights.”).
1262
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
A useful way to begin the discussion is to view the problem of
existing uses through the lens of retroactivity where a well-developed
body of scholarship has fleshed out closely related issues. The next
Section takes on this task. The remainder of this Part then considers
and rejects possible justifications—the particular character of the loss,
the preservation of the status quo, and the prevention of political process failures—for the protection of existing uses.
A. Antiretroactivity and Existing Uses
Many scholars argue that retroactive laws are anathema to liberty
and a well-ordered society.195 On the other hand, even the strongest
attacks against retroactive legislation recognize that the government
must have some ability to enact laws that affect settled expectations in
order to be able to legislate at all.196 Rules limiting retroactivity are
therefore in direct tension with the government’s ability to change the
law. As Daniel Troy has framed the problem: “To what extent is one
willing to sacrifice reliance on existing rules to accommodate the need
for change?”197
Much of the literature on retroactive laws focuses principally on
the surprisingly difficult question: What counts as a retroactive
law?198 This definitional problem is easy to identify but very difficult
195 See, e.g., LON L. FULLER, THE MORALITY OF LAW 53 (rev. ed. 1969) (“[A] retroactive law is truly a monstrosity.”); DANIEL E. TROY, RETROACTIVE LEGISLATION 17–19
(1998) (arguing that retroactive legislation fails to give notice and therefore undermines
free will and reliance necessary for social order); Julian N. Eule, Temporal Limits on the
Legislative Mandate: Entrenchment and Retroactivity, 1987 AM. B. FOUND. RES. J. 379, 427
(referring to common assumption that “retroactive exercises of legislative power are
incompatible with any fair and just system of government”).
196 See FULLER, supra note 195, at 60 (“If every time a man relied on existing law in
arranging his affairs, he were made secure against any change in legal rules, the whole
body of our law would be ossified forever.”).
197 TROY, supra note 195, at 3 (internal quotation marks omitted).
198 There are a variety of constitutional sources of antiretroactivity rules that apply to
private law. The Contracts Clause prevents state government from interfering with
existing contract rights (although often anemically). U.S. CONST. art. I, § 10, cl. 1 (“No
State shall . . . pass any . . . Law impairing the Obligation of Contracts . . . .”). Substantive
due process analysis replaced an early natural rights treatment of retroactive legislation.
TROY, supra note 195, at 77 (noting that by 1880s, Supreme Court had begun using substantive due process “to implement many of its conceptions of natural justice”). Courts
therefore reasoned that retroactive regulations, often described as those affecting vested
rights, were unconstitutional under the Due Process Clause. See Comment, The Variable
Quality of a Vested Right, 34 YALE L.J. 303, 304 (1925) (summarizing due process cases
and citing, inter alia, Huffman v. Alderson’s Adm’r, 9 W. Va. 616 (1876)). In 1976, however, the Supreme Court refused to strike down an expressly retroactive law on due process grounds. See Usery v. Turner Elkhorn Mining Co., 428 U.S. 1 (1976), cited in TROY,
supra note 195, at 38, 107 n.84. Nevertheless, substantive due process analysis still appears
periodically in retroactivity analysis. For example, in Eastern Enters. v. Apfel, 524 U.S. 498
(1998), a plurality of the Supreme Court struck down a federal statute requiring coal com-
November 2009]
EXISTING USES
1263
to solve. What counts as a retroactive law, when almost all laws alter
the consequences of past conduct?199 As Professor Jill Fisch has
explained, even prospective-only regulations can have a significant
retroactive effect.200 While certain laws may be more retroactive than
others, almost all affect prior conduct to some degree.201
Building on this insight, some scholars distinguish between
strongly and weakly retroactive laws.202 The former change a legal
panies to fund retirement benefits for former coal miners. The Court could not agree on
the constitutional analysis to apply. Four Justices believed the statute violated the Takings
Clause. Id. at 522–38 (plurality opinion) (O’Connor, J., joined by Rehnquist, C.J., and
Scalia and Thomas, JJ.). Justice Thomas suggested it also might violate the Ex Post Facto
Clause. Id. at 538–39 (Thomas, J., concurring). Justice Kennedy believed the statute was
unconstitutional but applied a due process analysis to reach his result. Id. at 547–50
(Kennedy, J., concurring in judgment and dissenting in part). The remaining four Justices
voted to uphold the statute after applying the Due Process Clause. Id. at 556–58, 567–68
(Breyer, J., dissenting). Given the Supreme Court’s fractured reasoning regarding retroactivity, it is perhaps not surprising that courts disagree about the source of the constitutional
protection for existing uses. For a review of different constitutional prohibitions against
retroactivity and their limitations, see Eule, supra note 195, at 427–34.
199 For example, if the government were to remove the home mortgage tax deduction it
would have an enormous impact on the value of people’s past investment decisions, even if
the deduction were only repealed prospectively. Jill E. Fisch, Retroactivity and Legal
Change: An Equilibrium Approach, 110 HARV. L. REV. 1055, 1067 (1997); see also Eule,
supra note 195, at 435–36 (considering same example). Tax law provides a frequent source
of discussion of retroactivity. See generally Kyle D. Logue, Tax Transitions, Opportunistic
Retroactivity, and the Benefits of Government Precommitment, 94 MICH. L. REV. 1129
(1996) (arguing for transition relief in certain tax situations).
200 Fisch, supra note 199, at 1067–69; see SHAVIRO, supra note 9, at 26 (identifying retroactive effect of “nominally prospective” changes); TROY, supra note 195, at 2 (“[A]lmost
all legislation may be characterized as ‘retroactive’ [because] the operation of almost all
legislation depends on antecedent facts.”); see also FULLER, supra note 195, at 59–61 (discussing “the most difficult problem of all, that of knowing when an enactment should properly be regarded as retrospective”). This identification problem is particularly serious for
land use regulations, because the land itself always preexists the regulation. See Holly
Doremus, Takings and Transitions, 19 J. LAND USE & ENVTL. L. 1, 11–12 (2003) (arguing
that “new property rules can never be wholly forward-looking,” because while “they can
be applied only to new activities, they can never be applied to new land”).
201 See Fisch, supra note 199, at 1070 (explaining that describing “retroactivity as a spectrum or range of temporal options rather than as a binary construct provides a better
description of the nature and consequences of legal change”); see also Logue, supra note
199, at 1133 (“Transition losses can occur whether the new tax law or new interpretation
applies nominally retroactively or nominally prospectively.”).
202 See, e.g., Stephen R. Munzer, Retroactive Law, 6 J. LEGAL STUD. 373, 383 (1977)
(explaining that retroactivity may be either strong or weak and that “the difference lies in
how the impact of a retroactive law on earlier acts in the period prior to its creation is
understood”); cf. TROY, supra note 195, at 8 (describing Justice Scalia as distinguishing
between primary and secondary retroactivity, which “roughly correlate” with strong and
weak retroactivity); W. David Slawson, Constitutional and Legislative Considerations in
Retroactive Lawmaking, 48 CAL. L. REV. 216, 216 (1960) (“[Retroactive] is used both (1)
to describe the particular basis of selection for the direct imposition of legal effects and (2)
as a description of particular kinds of effects which may occur when a new law is imposed
on society.”).
1264
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
status retroactively. The latter change a legal status prospectively only
but alter the consequences of events that predated enactment.203 For
an illustration in the land use context, imagine a new zoning ordinance
that eliminates a particular existing use. That zoning ordinance is not
strongly retroactive because it only operates prospectively—that is, it
requires only that people comply with the new law going forward and
does not impose civil or criminal penalties, say, for the existence of
the previously conforming use. It has a substantial “weak” retroactive
effect, however, insofar as it interferes with the owner’s prior investment in the existing use. But almost all land use regulations are retroactive in this sense. Any land use regulation has the potential to
undermine settled expectations around the uses of property.
To be clear, the retroactivity literature on its own does not
squarely address the problem of existing uses, let alone solve it. Even
a prospective land use regulation—one that does not implicate an
existing use—can have a retroactive effect in the sense that it interferes with the owner’s expectations about the use of her property.
Antiretroactivity rules, on their own, are therefore either under- or
overinclusive when it comes to protecting existing uses. Either they
only prohibit regulations that actually change preenactment legal statuses (i.e., that are strongly retroactive), permitting prospective regulations regardless of their impact on existing uses, or they prohibit
regulations that significantly interfere with settled expectations, in
which case they do not necessarily distinguish between regulations of
existing uses and regulations of undeveloped property where the
owner reasonably expected to be able to build.
The problem of defining retroactive legislation is difficult.
Harder still is the normative problem: deciding whether and to what
extent retroactive legislation should be permitted. According to many
accounts, some general sense of fairness requires that people be able
to order their affairs in reliance on a predictable set of rules.204 Retroactive laws have the effect of changing the rules in the middle of the
203
Munzer, supra note 202, at 383.
See Fisch, supra note 199, at 1084 (“It is typically thought that prospective laws are
more fair and that retroactive laws are more efficient.”); David Frisch, Rational Retroactivity in a Commercial Context, 58 ALA. L. REV. 765, 794 (2007) (“Traditional normative
criticism of retroactivity has rested on two related assertions: Fairness mandates giving
people the opportunity to know in advance what laws will govern their affairs and prohibits
changing the rules in midstream, and retroactive laws defeat the legitimate expectations of
the persons affected.” (internal footnotes omitted)); see also Landgraf v. USA Film Prods.,
511 U.S. 244, 265 (1994) (“[T]he presumption against retroactive legislation . . . embodies a
legal doctrine centuries older than our Republic. Elementary considerations of fairness
dictate that individuals should have an opportunity to know what the law is and to conform
their conduct accordingly; settled expectations should not be lightly disrupted.” (internal
footnote omitted)).
204
November 2009]
EXISTING USES
1265
game. Antipathy to retroactive laws is also based on the fundamental
requirement of notice and fair warning: People should be able to
know the legal status of their choices ahead of time.205
On the other hand, to the extent a new rule is an improvement to
the legal system, requiring it to apply only prospectively will limit the
rule’s benefits.206 The possibility of prospective application of an
unfavorable legal change may induce people to race to fall within the
existing regime.207 In the land use context, as explored in detail in
Part IV, this can lead to all kinds of inefficiencies as people engage in
strategic behavior to avoid application of a new rule by developing
property or engaging in transactions before they otherwise would.208
The question now squarely in focus is whether there is something
different about existing uses so that fairness- or efficiency-based arguments can justify their categorical protection. The remainder of this
Part examines the most likely justifications: that the extent and
nature of the loss associated with the elimination of an existing use is
greater and different in kind than the loss that results from the prohibition of a future use; that categorical protection of existing uses is a
form of protection for the status quo inherent in the nature of property rights; and that the political economy of existing uses demands
their protection. All three justifications turn out to be surprisingly
weak.
B. Nature and Extent of the Loss
Why do existing uses seem so different from future uses? Is there
any justification for courts treating them differently? There are a
number of ways to characterize the different nature of the harm
205 See TROY, supra note 195, at 18–19 (arguing that notice is fundamental to fairness,
rule of law, and moral choice); see also Fisch, supra note 199, at 1085 (“Notice enables
people to predict the consequences of their transactions and increases the influence of
legal rules upon primary conduct.”).
206 See Fisch, supra note 199, at 1088 (“The view that the new rule improves the operative legal principles supports the application of that rule to as broad a class of cases as
possible.”).
207 Id. at 1089 (“The objectives of a new legal rule may also be undercut if people are
able to avoid its application by rushing to complete transactions prior to enactment.”).
This is a familiar application of moral hazard and is considered in more detail in Part IV.A,
infra. See Baker, supra note 40, at 239 (“‘[M]oral hazard’ refers to the tendency for insurance against loss to reduce incentives to prevent or minimize the cost of loss.”); Albert C.
Lin, Size Matters: Regulating Nanotechnology, 31 HARV. ENVTL. L. REV. 349, 401 (2007)
(defining moral hazard as “the temptation to behave differently because of the existence of
insurance or other incentives”).
208 See infra Part IV.A; see also Kaplow, supra note 9, at 529 (“[T]he encouragement
resulting from the assurance that compensation or other protection will be provided in the
event of change results in overinvestment.”).
1266
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
resulting from the elimination of an existing use, but none are particularly satisfying.
1. Fairness
There seems to be something particularly unfair about regulating
away an existing use. As in the retroactivity literature, the fairness
concern is with the government changing regulatory rules on which
property owners have reasonably relied.209 Imagine someone who
developed a small commercial property only to have a new zoning
ordinance downzone the property exclusively for residential use. The
new ordinance interferes with the owner’s expectations about the use
of her property: She designed, built, and budgeted for a project based
on existing regulations that, once changed, dramatically reduce the
property’s economic value. This smacks of unfairness. Indeed, vested
rights and, potentially, zoning estoppel might preclude application of
the zoning change to the property, presumably to vindicate some combination of reliance interests and fairness.210
Notice, however, that this example is not conceptually distinct
from the example of an owner of undeveloped land who faces a
zoning change before having developed her property. In other words,
a prospective zoning change can also make a substantial investment in
undeveloped property retroactively unappealing and unaffordable. A
builder who spends a significant amount of money purchasing undeveloped property in reliance on existing zoning regulations that would
permit construction of commercial property presents the same reliance and fairness concerns when a zoning change leaves her with
property that can only be developed for less valuable residential use.
For the reliance interest to be especially applicable to the protection
209 See Doremus, supra note 200, at 14 (“Changing the rules after people have adjusted
their conduct on the basis of those rules often seems unfair, because we generally think
that people are entitled to, and indeed should, govern their behavior according to the
existing rules.”); Frisch, supra note 204, at 765–66 (describing intuition that “there is something fundamentally unfair about government altering the rules that govern past conduct”);
Matthew P. Harrington, Foreword: The Dual Dichotomy of Retroactive Lawmaking, 3
ROGER WILLIAMS U. L. REV. 19, 19 (1997) (“On the face of it, there is something unsettling about the possibility that lawmakers might change the rules whilst the game is in
progress.”). In a leading article, Joseph Singer offered an entire reliance-based theory of
property law. See generally Joseph William Singer, The Reliance Interest in Property, 40
STAN. L. REV. 611 (1988). It may be important to ask whether theories of reliance actually
have any place in property law, but that question is far beyond the scope of this project.
210 See, e.g., Kupchak et al., supra note 115, at 24 (“[T]he policy underlying zoning
estoppel is two-fold: hold the government to its commitments, and treat property owners
who rely fairly.”). Zoning estoppel can be applied to prevent the government from
changing zoning regulations on which a property owner has reasonably relied, but only if
the government has given assurances that the zoning will not change. Id.
November 2009]
EXISTING USES
1267
of existing uses, there must be something unique or at least particularly compelling about reliance that takes the form of an existing use.
Merely invoking reliance without providing this additional content
does not offer a basis for distinguishing between existing and future
uses.
2. Extent of Loss
One obvious distinction in the example above is the extent of the
property owner’s investment in her property and therefore the extent
of the loss she suffers when the zoning changes. This, however, is an
artifact of the specific example. In fact, viewed through a purely economic lens, the losses associated with regulating existing uses are not
necessarily greater than the losses associated with regulating prospective future uses.211 True, existing uses may increase the extent of loss.
All else being equal, eliminating an existing use is likely to impose a
more substantial loss than forbidding the same use prospectively. But
all else is not equal. Preventing a developer from developing a large
residential development (proscribing a future use) may impose much
more harm than requiring a property owner to tear down a shack
(eliminating an existing use).212 More fundamentally, if the extent of
economic loss is all that drives the protection of property, then
existing uses are irrelevant except as a crude proxy for loss which can
presumably be measured much more directly.
Of course, while the losses resulting from the elimination of an
existing use are not necessarily greater than the losses resulting from
the prohibition of a future use, they do have a different character.213
The former are out-of-pocket costs, the latter are foregone profits,
and people may experience these losses differently.214 This distinction
at least implicates a robust behavioral economics literature on endowment effects, which demonstrates that people generally value property
211 See Michelman, supra note 154, at 1233 (“[A] ban on potential uses not yet established may destroy market value as effectively as does a ban on activity already in
progress.”).
212 This kind of example is common in the case law. See, e.g., Savvidis v. City of
Norwalk, No. FSTCV054004143S, 2007 WL 2938522, at *4 (Conn. Super. Ct. Aug. 8, 2007)
(granting writ of mandamus on behalf of property owners that spent significant sums on
development in reliance on approval of zoning application).
213 Cf. Peter Huber, The Old-New Division in Risk Regulation, 69 VA. L. REV. 1025,
1065 (1983) (contrasting consequences of regulation of new products versus regulation of
existing products).
214 See SHAVIRO, supra note 9, at 23 (“People tend to over-weight out-of-pocket costs
relative to pure opportunity costs such as foregone gains. . . . Accordingly, out-of-pocket
losses and those that result in a perceived transaction loss may tend to be more salient than
others . . . .”).
1268
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
they actually possess more than property they do not.215 Simply put,
the owner of an existing use is likely to value it more than an otherwise identical use that does not yet exist.216 Moreover, experimental
evidence suggests that inchoate legal entitlements—like the right to
possess something in the future—generate weaker endowment effects
than the possession of actual, physical objects.217 This is consistent
with the intuition that actual, existing uses deserve more protection
than mere development rights or expectations. Also, endowment
effects appear to increase over time, reflecting the sense that the
longer a use has existed, the more protection it should receive.218
This powerful intuition does not have to be cast in behavioral
economics terms. Eliminating an existing use systemically imposes a
greater harm than prohibiting a prospective future use because of
people’s psychological connection with their property. As Justice
Holmes famously wrote about adverse possession, “A thing which you
have enjoyed and used as your own for a long time, whether property
or an opinion, takes root in your being and cannot be torn away
215 See Russell Korobkin, The Endowment Effect and Legal Analysis, 97 NW. U. L. REV.
1227, 1228 (2003) (“[P]eople tend to value goods more when they own them than when
they do not.”). See generally SHAVIRO, supra note 9, at 23 (discussing endowment effects);
Daniel Kahneman et al., Experimental Tests of the Endowment Effect and the Coase Theorem, 98 J. POL. ECON. 1325 (1990) (demonstrating experimentally that endowment effect
persists even in market setting with opportunities to learn).
216 See Michael Heller & Rick Hills, Land Assembly Districts, 121 HARV. L. REV. 1465,
1479–80 (2008) (discussing endowment effects and noting that “[h]umans have a wellverified psychological inclination to value their current endowments more than identical
items that they currently lack but could purchase”); Kahneman et al., supra note 215
(making same point).
217 See Jeffrey J. Rachlinski & Forest Jourden, Remedies and the Psychology of Ownership, 51 VAND. L. REV. 1541, 1558 (1998) (“[S]ubjects actually have to feel and touch [the
object] to make it theirs—the right to [an object is] not endowed.”); see also id. at 1559
(describing further psychological study); Doremus, supra note 200, at 23 (“The [endowment] effect does not seem to attach to expectations. The right to collect a commodity
does not give as strong an effect as even brief possession of the commodity itself . . . .”).
218 See Michael A. Strahilevitz & George Lowenstein, The Effect of Ownership History
on the Valuation of Objects, 25 J. OF CONSUMER RESEARCH 276 (1998) (publishing findings
from studies showing endowment effect increasing over time). Responding to this same
phenomenon, some have proposed increasing compensation for condemnation the longer
property has been owned by the condemnee. See, e.g., MO. ANN. STAT. § 523.039(3)
(West. 2009) (offering higher compensation for properties held by same family for fifty or
more years), cited in Daphna Lewinsohn-Zamir, Identifying Intense Preferences, 94
CORNELL L. REV. 1391, 1409 n.88; John Fee, Eminent Domain and the Sanctity of the
Home, 81 NOTRE DAME L. REV. 783, 814–17 (2006) (“For example, eminent domain law
might require government to pay homeowners market value plus X percent of the home’s
market value, where X depends on how long the owner has lived in the home.”). It may,
however, be difficult to move from controlled psychological experiments involving tangible
and low-value property to real property. See Jennifer Arlen, Comment, The Future of
Behavioral Economic Analysis of Law, 51 VAND. L. REV. 1765, 1768–71 (1998) (“The
classic experiment illustrating the endowment effect involved Cornell coffee mugs.”).
November 2009]
EXISTING USES
1269
without your resenting the act and trying to defend yourself.”219 This
sentiment is equally applicable to the protection of existing uses. It is
hard to deny the entitlement that people feel to continue a use that
already exists.220 As Judge Henry Friendly once wrote, discussing
procedural due process, “[W]hatever the mathematics, there is a
human difference between losing what one has and not getting what
one wants.”221 However framed, people’s connection to their property might justify existing use protection because the character of the
relevant harm is different.
While this account may have considerable descriptive force, it is
difficult to justify treating these psychological connections differently
from property owners’ subjective valuation of their property,
whatever the source. It is well-settled doctrine that when the government takes property, owners are entitled to its fair market value and
do not receive compensation for subjective value.222 Courts have reasoned that the difficulty—indeed, the seeming impossibility—of measuring subjective value with any precision militates in favor of
ignoring it altogether.223 There is really no conceptual difference
between the value resulting from the actual possession of property
and subjective value resulting from, say, property being in the family
for a long time. Both are forms of the subjective value normally
excluded from takings protection.
The comparison to the broader category of subjective value
reveals yet another problem. The psychological connections between
people and their property implicate the extent of harm that results
from the elimination of an existing use. Most arguments about
including subjective value in takings analysis therefore focus on providing adequate compensation. A variety of academic proposals to
219
O.W. Holmes, The Path of the Law, 10 HARV. L. REV. 457, 477 (1897).
In support of this intuition, Daniel Troy cites a text on child development that
stresses the importance of knowing applicable rules ahead of time. See TROY, supra note
195, at 1 (citing H. CLAY TRUMBULL, HINTS ON CHILD TRAINING 216 (1890)).
221 Henry J. Friendly, “Some Kind of Hearing,” 123 U. PA. L. REV. 1267, 1296 (1975),
cited in Greenholtz v. Inmates of the Neb. Penal & Corr. Complex, 442 U.S. 1, 10 (1979).
Presaging the economic discussion that follows, Friendly continued, “Revocation of a
license is far more serious than denial of an application for one; in the former instance
capital has been expended, investor expectations have been aroused, and people have been
employed.” Id. For further discussion, see Susan H. Herman, The New Liberty: The Procedural Due Process Rights of Prisoners and Others Under the Burger Court, 59 N.Y.U. L.
REV. 482, 514 (1984).
222 Christopher Serkin, The Meaning of Value: Assessing Just Compensation for Regulatory Takings, 99 NW. U. L. REV. 677, 700–01 (2005).
223 See Id., at 701 n.110 (citing cases); see also Lee Anne Fennell, Taking Eminent
Domain Apart, 2004 MICH. ST. L. REV. 957, 993–95 (“It is difficult to know how much
value someone places on a property, and resort to proxies such as percentage bonuses will
generate inaccuracies in both directions.”).
220
1270
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
address the problem of subjective value—from a self-valuation model
offered by various scholars to awarding percentage bonuses—each
increase the amount of compensation the government must pay but
do not affect the initial determination of whether the government
must pay.224 There is no obvious reason to treat existing uses any differently. If psychological attachment to existing uses implicates the
extent of harm but not any qualitative or conceptual differences with
prospective uses, then they are better addressed in the compensation
inquiry than in the liability determination. But this is not how current
land use and property doctrines work.225
Finally, the fact of an existing use is at best a rough proxy for the
kinds of psychological attachments that the law might reasonably seek
to protect, administrability concerns notwithstanding. Individuals,
and homeowners in particular, may experience endowment effects
that change their relationship to existing uses. Applying that intuition
to justify protecting commercial property—not to mention more
abstract investments in property—is more of a stretch.226
3. Waste
Perhaps, though, even if the magnitude of the loss cannot justify
existing use protection, the character of the loss can. A regulation
eliminating an existing use wipes out the money already expended in
developing that use. This may seem like a qualitatively different kind
of harm because it appears to eliminate a productive and durable
asset, thereby imposing some broader cost on society.227 In fact, however, when it comes to protecting existing uses, this intuition relies on
the same unwarranted prioritization of out-of-pocket costs over foregone profits.228
224 For a proposal for self-assessment of property values tied to property taxes and sales
prices, see Abraham Bell & Gideon Parchomovsky, Takings Reassessed, 87 VA. L. REV.
277, 300–04 (2001). Richard Epstein has suggested applying some extra percentage
increase to compensation at least partly to account for subjective value. Richard A.
Epstein, A Clear View of The Cathedral: The Dominance of Property Rules, 106 YALE L.J.
2091, 2114–15 (1997). For a discussion of the possibility of self-assessment in other contexts, see generally Saul Levmore, Self-Assessed Valuation Systems for Tort and Other Law,
68 VA. L. REV. 771 (1982).
225 See supra Part I (describing property and land use doctrines protecting existing
uses).
226 See infra Part III.B.4 (discussing personhood).
227 See Dana, supra note 40, at 685 (“The costs of undoing an existing development,
moreover, are typically much greater than the costs of preventing development.”); cf. Lior
Jacob Strahilevitz, The Right To Destroy, 114 YALE L.J. 781, 796 (2005) (“Concern about
wasting valuable resources is, by far, the most commonly voiced justification for restricting
an owner’s ability to destroy her property.”).
228 See supra note 214 (discussing such overweighting). This also resembles at least
some characterizations of the sunk cost fallacy in that it unduly privileges money that has
November 2009]
EXISTING USES
1271
To see why, consider the following stylized hypothetical. Imagine
that Los Angeles is interested in zoning some property exclusively for
low–income housing and can choose between two lots. The first, in a
mixed–income neighborhood, already has a movie theater on it.
Zoning that property would render the theater unusable and would
further reduce the value of the property for a combined loss of, say,
five hundred thousand dollars.229 The second is an undeveloped lot in
the toniest part of Beverly Hills. Zoning it for low-income housing
will reduce the value of the property by five million dollars. All else
being equal, there is no doubt that the government should choose to
zone the movie theater out of existence instead of downzoning the lot
in Beverly Hills. The supposed waste resulting from the destruction of
the theater is fully captured by the use value of the property. The fact
that an existing movie theater is eliminated does not create some additional harm to society above and beyond what is captured by its present market value (subject only to endowment effects and the like).230
However, eliminating the movie theater is likely to be a taking while
downzoning the undeveloped property is not.
Professor Shavell has recently modeled one form of this more
general problem, demonstrating that transition costs for modifying
durable uses, or otherwise eliminating nonconforming uses, need to be
included in the cost-benefit analysis to decide whether to grandfather
them.231 His model reveals the potential significance to transition
costs of investments in existing uses (what he calls “durable precautions”) but simultaneously recognizes that grandfathering can be
already been spent. See Susan Block-Lieb & Edward J. Janger, The Myth of the Rational
Borrower: Rationality, Behavioralism, and the Misguided “Reform” of Bankruptcy Law, 84
TEX. L. REV. 1481, 1534 (2006) (“Cognitive research also finds that individuals are reluctant to walk away from sunk costs, irrationally ignoring the marginal costs and benefits of
additional action.”); Mark Seidenfeld, Cognitive Loafing, Social Conformity, and Judicial
Review of Agency Rulemaking, 87 CORNELL L. REV. 486, 500 (2002) (“When making
investment decisions, the old economic adage advises that one should ignore sunk costs,
but people intuitively tend to let such costs influence their choices.”).
229 That loss includes both the costs of tearing down the theater and the lesser value of
the property as low-income housing in relation to the higher-valued movie theater.
230 For a discussion of endowment effects, see supra notes 216–18 and accompanying
text. This waste analysis will not hold true if eliminating the existing use imposes an
intergenerational externality. See Strahilevitz, supra note 227, at 793–94 (identifying
problem with property destruction as imposing “intergenerational consequences, for better
or worse”). By the same token, a regulation can also impose an intergenerational benefit.
There is no reason, in the abstract, to suspect that existing uses will be more valuable in the
future than the kinds of regulatory benefits the government might seek to obtain by eliminating an existing use. Unless there is some kind of systemic valuation failure for existing
uses, the present value of existing uses should capture their value in the future.
231 Shavell, supra note 17, at 57–64.
1272
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
unnecessarily costly.232 In other words, and consistent with the central
claim here, while the fact of some kinds of preexisting uses of property
can be relevant to the efficiency of new regulations, it plainly is not
dispositive and so does not justify categorical protection.
The picture becomes even more complex if the problem of waste
is viewed from a different temporal perspective. Instead of focusing
purely on the government’s decision whether to regulate after a use is
already in place (i.e., after costs have already been sunk), it is more
useful to ask what rules should be in place to minimize the opportunity for waste ex ante. Of course, both property owners and the government have the chance to prevent waste: the property owner by not
building in the first place; the government by not regulating after the
use already exists. Under one standard law and economics account,
the risk of loss should be placed on the party best situated to avoid
it.233 In other words, waste will be minimized if the least-cost avoider
bears the risk of a use being eliminated.
This approach does not compel categorical protection of existing
uses. While the government always has the last chance to avoid the
loss—the government is always in a position not to regulate—
property owners may frequently be in a better position to minimize
costs more cheaply by not building in the first place.234 There are, in
fact, strong reasons to think that compensation may result in overinvestment in property, a point taken up in Part IV.235 At the very
least, the fact-specific and empirical question of who the least-cost
avoider is in any particular situation cannot justify protection for
existing uses in all or even most cases.236
232 See id. at 47 (“[A] party ought to continue with its period 1 precaution in period 2 if
the cost of the new conventionally optimal precaution for period 2 harm would exceed the
marginal reduction in expected harm that would be accomplished by a change to this
precaution.”).
233 See, e.g., Scott Hershovitz, Two Models of Tort (and Takings), 92 VA. L. REV. 1147,
1152 (2006) (“[T]he law aims to place liability on the least-cost avoider. The least-cost
avoider is frequently the party best able to control a situation.”); Eric Rasmusen, Agency
Law and Contract Formation, 6 AM. L. & ECON. REV. 369, 380 (2004) (“The least-cost
avoider principle, broadly stated, asks which party has the lower cost of avoiding harm, and
assigns liability to that party.”).
234 This kind of reasoning is reminiscent of the “last clear chance” doctrine in torts,
which “allowed plaintiffs to prevail despite their being the cheaper precaution taker
against their injury.” Robert H. Heidt, When Plaintiffs Are Premium Planners for Their
Injuries: A Fresh Look at the Fireman’s Rule, 82 IND. L.J. 745, 771 (2007). This doctrine
has been replaced in most jurisdictions by comparative fault regimes. Ehud Guttel, The
Hidden Risk of Opportunistic Precautions, 93 VA. L. REV. 1389, 1418 n.62 (2007).
235 See infra notes 278–90 and accompanying text (discussing transitions literature).
236 For examination of the complexities of identifying the least-cost avoider in another
context, see Oren Bar-Gill, The Behavioral Economics of Consumer Contracts, 92 MINN.
November 2009]
EXISTING USES
1273
4. Personhood
It may be that these economics-oriented accounts miss some
more fundamental aspects of existing uses and the connection that can
develop between people and their property. As Professor Radin has
observed, some relationships between people and their property
implicate the owner’s personhood.237 When this is the case, according
to Radin, compensation not only fails to provide sufficient protection
but also threatens to work greater injury by commodifying what
should be uncommodifiable.238
Existing uses implicate and simultaneously expand upon this personality theory of property. In Radin’s view, property can be constitutive of identity: The relationships between persons and external
objects are what create fully developed individuals.239 These relationships, however, rely on actual objects in the world and not on a mere
expectancy or hope for objects in the future.240 They rely, in other
words, on existing uses and not merely prospective future ones.
Existing uses are at most a necessary but not sufficient precondition for property to implicate personhood. Radin identifies a continuum from personal to fungible property: The former is constitutive
of identity, while the latter is held purely instrumentally.241 A prospective future use is, in this sense, always farther to the fungible side
of the property spectrum than an existing use. As Radin herself
explains, “Object-loss is more important than wealth loss because
object-loss is specially related to personhood in a way that wealth-loss
is not.”242 But it is not the case that all or even most existing uses
implicate owners’ personhood at all, let alone to an extent that
L. REV. 749, 791 (2008), which discusses the asymmetrical positions of consumers and
sophisticated sellers with regard to the ability to avoid mistakes.
237 See Margaret Jane Radin, Property and Personhood, 34 STAN. L. REV. 957, 965–66
(1982) (discussing whether people have property interest in their bodies).
238 See Margaret Jane Radin, Market-Inalienability, 100 HARV. L. REV. 1849, 1907–09
(1984) (describing application of market-inalienability theory to personhood).
239 Radin, supra note 237, at 972–75 (discussing Hegelian theory).
240 As Radin describes it, “Hegel’s property theory is an occupancy theory; the owner’s
will must be present in the object.” Id. at 973. The primacy of the object’s existence, then,
is essential to Radin’s personhood perspective on property. Extrapolating from the
“notion that the will is embodied in things,” Radin concludes that “[t]he idea of embodied
will . . . reminds us that people and things have ongoing relationships which have their own
ebb and flow, and that these relationships can be very close to a person’s center and
sanity.” Id. at 977. But see Radin, supra note 161, at 1692 (suggesting people may have
personhood interests in plans to build residence).
241 Radin, supra note 237, at 959–60 (describing fungible property as opposite of personal property). Radin argues that “[t]he opposite of holding an object that has become a
part of oneself is holding an object that is perfectly replaceable with other goods of equal
market value. One holds such an object for purely instrumental reasons.” Id.
242 Id. at 1004.
1274
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
demands protection. The kinds of existing uses frequently at issue in
land use disputes are commercial uses, or property owned by corporations. Personhood in such contexts simply is not—or at least not
often—at issue.243
C. Protecting the Status Quo
1. Objective Stability
Another kind of explanation for the protection of existing uses
may come from property law’s more general role in promoting stability.244 Indeed, property’s long-established concern with perpetuating the status quo has occasionally led to its being cast as an
ossifying force and even an enemy of progress.245 For better or worse,
property’s protection of existing uses may simply be another expression of its general resistance to change. While this characterization of
property’s stabilizing character has some appeal, it does not withstand
serious scrutiny as either a positive or normative account of existing
use protection.
First, descriptively, the stability that property promotes may be
either stability in the regulatory regime or stability in the actual uses
of property. The former does not provide any basis for distinguishing
between existing and future uses and is easily set aside. It is simply
243 Individual owners of small businesses may well have their identities bound up in
their commercial property. For a pleasant example, see Donna Paul, Fashion’s Just a Job;
Baking’s a Destiny, N.Y. TIMES, Sept. 28, 2007, at F3 (describing young couple buying
bakery that appeared far more important to them than their house). Line drawing is nevertheless hard, if not impossible. In an interesting new article, Professor Stephanie Stern
questions whether personhood interests are actually at stake in the home or whether “residential protectionism” is simply the result of interest group capture. See Stephanie M.
Stern, Residential Protectionism and the Legal Mythology of Home, 107 MICH. L. REV.
1093, 1096 (2009).
244 See, e.g., Carol M. Rose, The Shadow of The Cathedral, 106 YALE L.J. 2175, 2187
(1997) (“The usual roles of property rules—defining rights and identifying rights-holders—
. . . encourage individual investment, planning, and effort, because actors have a clearer
sense of what they are getting.”); see also Abraham Bell & Gideon Parchomovsky, A
Theory of Property, 90 CORNELL L. REV. 531, 573 (2005) (“The stability in ownership
afforded by the law creates the possibility for developing new kinds of value in, and uses
of, property that would otherwise be unavailable.”); Thomas W. Merrill & Henry E. Smith,
Optimal Standardization in the Law of Property: The Numerus Clausus Principle, 110
YALE L.J. 1, 64 (2000) (“A rule that has been around a long time and is relatively
unchanging is more likely to be understood because actors . . . are more apt to have
encountered the rule in the past and to have made some previous investment in comprehending the rule.”).
245 See, e.g., Craig Anthony (Tony) Arnold, The Reconstitution of Property: Property as
a Web of Interests, 26 HARV. ENVTL. L. REV. 281, 325 (2002) (“[Property] may involve a
‘double bind’ between reinforcing the status quo by compromising ideals and making very
little real progress toward change, and reinforcing the status quo by insisting on utopian
ideals and making no progress toward change.”).
November 2009]
EXISTING USES
1275
about minimizing regulatory change and is indifferent to the use of
property. The latter is descriptively inadequate as a justification for
existing use protection. The now familiar—if not standard—account
of property law is organized around creating incentives to put property to more productive use.246 Indeed, the history of property law in
this country is the history of promoting increasingly intensive uses of
land, of chattles, of ideas. The guiding principle has not been maintaining stability but rather encouraging productivity. Existing use protection is only consistent with this history to the extent that it protects
existing uses against less intensive uses and not more intensive ones.
This distinction simply does not exist in the doctrine.
2. Evidence of Subjective Expectations
Instead of protecting some kind of objective stability, perhaps
existing use protection can be explained as protecting property
owners’ subjective expectations about the use of property. The idea is
simply that property owners should be protected from regulatory
changes that interfere with real and reasonable expectations. Existing
uses may be useful for deciding what count as subjectively real and
objectively reasonable expectations because they serve as definitive
evidence of some baseline of actual expectations concerning the use of
property.247
Whether a use exists or not has little to do with its objective reasonableness; plenty of unreasonable land uses actually exist.248 But
existing uses might be useful for converting inscrutable individual
plans into protectable property interests. It is one thing to protect a
property owner’s claim that she hoped, someday, to develop her property. It is, perhaps, something else to protect the property owner who
has already actually done so. Viewed this way, existing uses have salience primarily because they are particularly good evidence of a property owner’s real expectations about the use of property.
246 See, e.g., Richard A. Epstein, International News Service v. Associated Press:
Custom and Law as Sources of Property Rights in News, 78 VA. L. REV. 85, 108 (1992)
(arguing that distribution of property rights is “functional” and “forward-looking” and suggesting that “the central task is to develop that initial distribution of rights that leads to the
shortest path for the productive use of natural resources—that is, to some form of allocative efficiency”).
247 This may also be what Michelman was suggesting in the quotation accompanying
note 162, supra.
248 The harm-prevention exception for regulatory takings liability clearly demonstrates
that property is regularly put to a use that is, or later turns out to be, unreasonable. See,
e.g., Lucas v. S.C. Coastal Council, 505 U.S. 1003, 1026 (1992) (articulating nuisanceprevention defense); Hadacheck v. Sebastian, 239 U.S. 394 (1915) (upholding law barring
operation of brick mills in residential areas).
1276
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
Of course, existing uses are only evidence of some floor of expectations, not the ceiling. Property protection does not extend only to
existing uses. Moreover, existing uses are not the only good evidence
of an owner’s expectations. Where an owner claims an expectation
beyond the existing use of the property, submitted proposals, architect
renderings, and even the purchase price of undeveloped property can
all demonstrate the sincerity of those plans. This kind of inquiry is
entirely consistent with the vested rights doctrine249 as well as with
Penn Central’s focus on investment-backed expectations.250 Still,
existing uses seem like the best evidence of actual expectations about
the use of property, expectations that the government cannot abrogate without paying compensation.
This account has some descriptive power, but it ultimately does
not justify special protection for existing uses. Most profoundly, it
does not explain why a property owner’s individual plans should be a
relevant consideration for property protection in the first place. If the
concern is distinguishing pie-in-the-sky development dreams from
real, concrete plans, diminution in value already captures this, without
focusing at all on actual expenditures or existing uses, or even on subjective expectations. The more speculative and unrealistic an expectation, the less its elimination will impact fair market value.251 That is,
market value will not reflect unreasonable expectations.
If, instead, the concern is genuinely with protecting subjective
expectations—regardless of the impact on the property’s fair market
value—the result is simply inconsistent with core takings doctrine.
When it comes to prospective future uses, the law does not even
aspire to protect all genuine and reasonable expectations.252 Developers are routinely required to scale back projects and offer concessions as part of the development process regardless of the sincerity of
249
See supra Part I.C (discussing vested rights doctrine).
See supra notes 132–56 and accompanying text (discussing investment-backed expectations prong of Penn Central test).
251 See Serkin, supra note 222, at 689–92 (discussing highest and best use as component
of fair market value).
252 See Katrina Miriam Wyman, The Measure of Just Compensation, 41 U.C. DAVIS L.
REV. 239, 250–51 (2007) (problematizing what it means for compensation to make property owners whole).
250
November 2009]
EXISTING USES
1277
their original expectations.253 So long as that interference does not go
too far, it is not impermissible and does not trigger takings liability.254
Protecting existing uses because they are evidence of some baseline of a property owner’s expectations unjustifiably presumes that an
existing use also defines the limits of the government’s ability to interfere with those expectations. To see the problem, imagine a developer
who, in negotiations with a local government, is forced to change her
plans for a condo development to build fifty-nine instead of sixty
units. It is hard to believe this demand could ever run afoul of the
Takings Clause, despite its interference with the owner’s sincere
expectations.255 But now imagine a local government seeking to force
the removal of one condo unit out of sixty that have already been
built. These two situations feel different—but not because there is
any difference in the extent of the regulations’ interference with the
property owner’s real expectations. Rather, the difference lies in the
magnitude of the loss, which the previous section has already rejected
as a legitimate basis for existing use protection.
In sum, then, neither a desire for stability (either of regulation or
of land use) nor an interest in protecting owner expectations can completely explain or justify a categorical rule protecting existing uses.
D. Political Economy
An entirely different kind of normative justification for existing
use protection comes not from the character of the loss, or protection
of the status quo, but instead from the potential for systemic political
malfunction. According to Professor Fischel, among others, property
protection should be at its highest when political protections are at
their weakest.256 The question, then, is whether owners of existing
253
Carol Rose offered the following examples of such tradeoffs and concessions:
A community might ask a developer to provide park space as a tradeoff for
permission to build a new development, or to preserve a familiar community
landmark in exchange for permission to build at a higher density; or the
tenants of a low income area, through the local government as mediator, may
negotiate with a highrise developer for low-income housing to offset the loss of
inexpensive residential hotel space.
Rose, supra note 46, at 891 (footnotes omitted).
254 For a discussion of the Penn Central factors, see supra Part II.B.
255 See supra note 158 and accompanying text (discussing diminution in value test for
takings liability).
256 WILLIAM A. FISCHEL, REGULATORY TAKINGS 139 (1995) [hereinafter FISCHEL, REGULATORY TAKINGS]. See generally WILLIAM A. FISCHEL, HOMEVOTER HYPOTHESIS (2001)
[hereinafter FISCHEL, HOMEVOTER HYPOTHESIS] (arguing that in face of homeowners’
political power, developers and owners of undeveloped property need greater protection
of property rights). Dean Treanor adopted a similar approach in his leading historical
treatment of the Takings Clause, arguing for a political process-based interpretation
focused on the political protection of property. Treanor, supra note 165, at 708–10; see also
1278
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
uses are likely to suffer from a predictable political failure. The
answer turns out to be somewhat complex.
In his examination of interest group politics, Professor Komesar
argues that the principal requirements for generating political pressure are relatively small numbers and high per capita stakes.257 The
size of the interest group matters because it affects the ability to identify interested parties as well as the cost of organizing them.258 The
per capita stakes matter because they increase the incentives of the
interested parties to expend the time and money necessary to overcome the organizational costs.259 These requirements neatly track the
characteristics of owners of existing uses, who not only constitute a
defined class but also are often easy to identify as the record owners
of apparent uses of land. The transaction costs required for their
coordination should therefore be relatively low. Moreover, the per
capita stakes are predictably high for all the reasons discussed
above.260 In the rough-and-tumble of interest group politics, the class
of existing use owners should be far less susceptible to the kinds of
political failures that public choice theory predicts when costs are
imposed on diffuse, disinterested, and unorganized groups.261
In contrast, prospective future uses are more susceptible to government interference than existing uses. Many of the beneficiaries of
Doremus, supra note 200, at 40 (“More searching review is appropriate where only a
minority will bear the regulatory burden . . . .”); Saul Levmore, Takings, Torts, and Special
Interests, 77 VA. L. REV. 1333, 1344–45 (1991) (“[W]hen . . . the government’s aims could
have been achieved in many ways but the means chosen placed losses on an individual or
on persons who are not part of an existing or easily organized political coalition, then we
can expect to find a compensable taking.”).
257 NEIL K. KOMESAR, LAW’S LIMITS 61 (2001) (explaining advantage of “groups with
small numbers but high per capita stakes,” whose members “have greater incentive to
expend the effort necessary to recognize and understand the issues” over “groups with
larger numbers and smaller per capita stakes”).
258 Id. (“[F]or larger groups, the cost of participation depends heavily on the cost of
organization, which in turn depends on both the size of the group to be organized and the
difficulty of identifying and convincing potential allies.”).
259 Id.
260 See supra Part III.C.2 (discussing subjective value).
261 See Einer R. Elhauge, Does Interest Group Theory Justify More Intrusive Judicial
Review?, 101 YALE L.J. 31, 37 (1991) (predicting that “small groups with concentrated
(high per capita) interests in lawmaking will come closer to their optimal level of petitioning than large groups with diffuse (low per capita) interests”); Amnon Lehavi & Amir
N. Licht, Eminent Domain, Inc., 107 COLUM. L. REV. 1704, 1719 (2007) (“[Public choice
theory] points to the disproportionate influence that special interest groups, and chiefly
politically powerful real estate entrepreneurs, have over governmental decisions in land
policy issues.”). For further discussion of the insights of public choice theory and its implications for property protection, see generally Daryl J. Levinson, Making Government Pay:
Markets, Politics, and the Allocation of Constitutional Costs, 67 U. CHI. L. REV. 345 (2000),
and Christopher Serkin, Big Differences for Small Governments: Local Governments and
the Takings Clause, 81 N.Y.U. L. REV. 1624, 1637–43 (2006).
November 2009]
EXISTING USES
1279
a proposed housing development, for example, are unidentified endpurchasers who are therefore unable to mobilize political opposition
to new regulation.262 The developer certainly has an interest—indeed
a strong interest—in resisting regulations limiting or eliminating a new
development, but even so, that developer is unlikely to internalize all
of the costs that the government restriction will impose.263 For these
and similar reasons, Professor Fischel has argued that the Takings
Clause should principally be concerned with protecting undeveloped
property because of the likely political failures surrounding local governments’ land use regulations.264
In fact, however, this political economy account is not so cut-anddried. Predictable political process failures in some contexts suggest a
particular need to protect existing uses from exploitation. Consider
Albert Hirschman’s concepts of “Exit” and “Voice.”265 Owners of
existing uses may well have a louder voice in the political process, but
it comes with an offsetting loss in the ability to exit. The threat of exit
is already low with undeveloped property. As Fischel observes, real
property cannot simply be moved.266 If the government passes a new
regulation severely limiting the use of property, the owner’s only realistic option is to cash out, often at a considerable loss. The costs associated with moving are an opportunity for rent-seeking by the
government; so long as the costs of moving are more than the costs
imposed by the government, a rational property owner will stay in the
262 See, e.g., Elhauge, supra note 261, at 39 (characterizing “[l]arge diffusely interested
groups” as underrepresented and as having “a harder time collecting the resources necessary to monitor and evaluate developing issues, make campaign contributions, [and] present information to voters or officials”).
263 For an account of those costs, see Serkin, supra note 261, at 1677–79.
264 FISCHEL, REGULATORY TAKINGS, supra note 256, at 139, 282; see Hanoch Dagan,
Takings and Distributive Justice, 85 VA. L. REV. 741, 777 n.119 (1999) (interpreting Fischel
as supporting judicial intervention to protect undeveloped land from local governments);
Carol M. Rose, Takings, Federalism, Norms, 105 YALE L.J. 1121, 1126 (1996) (book
review) (same); see also William A. Fischel, Exploring the Kozinski Paradox: Why Is More
Efficient Regulation a Taking of Property?, 67 CHI.-KENT L. REV. 865, 866 n.7 (1991) (“In
many cases, however, zoning imposes burdens on one set of owners (e.g., owners of undeveloped land) to benefit another set of owners (e.g., owners of already-developed
homes).”).
265 See ALBERT O. HIRSCHMAN, EXIT, VOICE & LOYALTY: RESPONSES TO DECLINE IN
FIRMS, ORGANIZATIONS, AND STATES 3 (1970). For a thoroughgoing account of the role of
exit and voice in local governments, including a description of Hirschman’s theory, see
Vicki Been, “Exit” as a Constraint on Land Use Exactions: Rethinking the Unconstitutional
Conditions Doctrine, 91 COLUM. L. REV. 473, 476 (1991).
266 See FISCHEL, REGULATORY TAKINGS, supra note 256, at 139 (“Owners of property
whose services are elastic in supply can protect themselves from myopic local regulation by
threatening to leave the jurisdiction. The remaining category, regulation of immobile
property by independent local governments and state agencies, requires most of the attention of judges in regulatory takings cases.”).
1280
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
jurisdiction.267 If the costs of exit for owners of undeveloped property
are high, they are often even higher for owners of developed property,
and when they are, the opportunity for exploitation by the government is higher too.268
Existing uses are also particularly salient, however, because of the
opportunity they present for the government to single out specific
property owners to bear regulatory burdens.269 There are times—
predictable times—when eliminating an existing use is likely to generate, not expend, political capital. Zoning away adult businesses is
perhaps the easiest example, in which the costs are imposed on the
business owner and the patrons who, because of the nature of the use,
are unlikely to mobilize to create coordinated political pressure. But
this same political dynamic exists whenever a municipality can single
out a politically disfavored use for the benefit of the majority or,
simply, the politically powerful.
The opportunity for this kind of singling out is likely to vary
depending on the nature and size of the government. It is more present in small, local governments where the stakeholders are clear and
the range of interests is relatively narrow. In small municipalities, primarily responsive to homeowner majorities, local politicians maximize
their political power by catering to homeowner interests.270 Individual homeowner preferences may vary, but homeowners as a group
share a common goal of maximizing property values.271 They are also
uniquely situated to exert political pressure on local politicians, both
because they constitute a majority of voters in most small jurisdictions
and because they enjoy informal organizational advantages.272
Homeowner interests with respect to existing uses are relatively
easy to anticipate. In general, homeowners will want their own homes
protected from regulatory change.273 They may feel differently, however, about nonresidential uses, especially those that have (or are perceived to have) an adverse impact on residential property values. If
this means shutting down a small truck loading facility or a nearby
267 For a discussion of rent-seeking in land use regulations, see Abraham Bell & Gideon
Parchomovsky, Compensating Against Corruption 3 (Mar. 3, 2008) (unpublished manuscript, on file with the New York University Law Review) (arguing takings compensation is
justified because it reduces profitability of rent-seeking behavior).
268 This claim comes with a substantial caveat: Some uses can be moved relatively
easily. For example, many businesses do not depend on a specific location.
269 Cf. SHAVIRO, supra note 9, at 77 (“[T]here plainly would be a strong argument for
allowing ex post facto criminal legislation, but for a concern about singling out.”).
270 FISCHEL, HOMEVOTER HYPOTHESIS, supra note 256, at 87–96 (developing median
voter model).
271 Serkin, supra note 261, at 1648.
272 Id.
273 FISCHEL, HOMEVOTER HYPOTHESIS, supra note 256, at 9 (discussing NIMBYism).
November 2009]
EXISTING USES
1281
stable with braying donkeys, so be it. In the absence of a compensation requirement, the government is very likely to regulate regardless
of the financial harm it is imposing on the owner of the existing use
because the political interests are so neatly aligned to favor
homeowners.
The same is simply not true in larger governments. There,
interest group politics play out far less predictably. The organizational advantages that owners of existing uses enjoy are not offset by
the coordinated political force of local homeowners.274 The political
failure to worry about in larger governments is their responsiveness to
owners of existing uses, not the opposite. While exceptions are easy
to imagine, in general the protection of existing uses therefore appears
to be more important for small local governments than for larger or
higher levels of government.275 Indeed, existing uses appear to fare
well—perhaps too well—at the state and federal level, as various state
and federal statutes demonstrate.276
Ultimately, then, it is difficult to generalize about the political
power of owners of existing uses. It is likely to depend on the nature
of the government and the nature of the use. This variety in the political story cannot justify the blanket protection that existing uses
receive. It is simply not the case that existing uses suffer from systemic political failures that justify greater judicial protection across
the board.
IV
LIMITING EXISTING USE PROTECTION
Despite its strength, the intuition that existing uses demand categorical protection turns out to be surprisingly difficult to defend. The
fundamental problem is that the presence of an existing use does not
always correspond to the plausible interests that existing use protection seeks to safeguard. Note that this is not an argument that existing
uses are irrelevant, or that they never represent important interests.
On the contrary, the presence of an existing use is often a good proxy
274 See Serkin, supra note 261, at 1661–64 (arguing that homeowners’ political power
dissipates at state and federal level).
275 The prescriptions arising from this likely political failure are discussed in more detail
in Part IV.C, infra. Here, it is enough to repeat Professor Fischel’s argument that in the
takings context, judicial resources should be focused only on those situations where property owners are unlikely to have a meaningful political remedy. See Fischel, supra note
264, at 911–12.
276 Principal among those statutes are the Clean Air Act, the Clean Water Act, the
Wilderness Act, the Americans with Disabilities Act, and a number of state statutes, each
of which provides special treatment for existing uses. See supra notes 12–24 and accompanying text.
1282
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
for high transition costs or high subjective value, to take just two
examples. The conclusion of the previous Part is simply that the categorical protection of existing uses extends more protection than
appropriate in at least some cases. The line between existing uses and
prospective future uses is not grounded in inherent differences
between the two.
It is therefore still worth exploring whether a categorical
approach to existing uses makes sense as a kind of prophylactic rule.
Instead of delineating some special class of property or entitlements,
the category of existing uses may be useful as a proxy for concerns like
the extent of loss, transition costs, or high subjective value. In this
way, categorical existing use protection can perhaps be defended as an
easy-to-administer bright-line rule. Although overbroad, it prevents
harms that are often, even if not always, associated with the elimination of an existing use. This protection may be justified after all if it
“gets it right” often enough, and at low enough cost, so that it is better
than the alternative.
Ultimately, this possible justification implicates an empirical
question about the relative costs and benefits of existing use protection. It also implicates an entire academic literature on the benefits of
bright-line rules, and the necessary fit between constitutional limits
and the interests being protected.277 Indeed, if existing use protection
were costless—or even just sufficiently cheap—an admittedly overbroad prophylactic rule might make good sense. It would minimize
uncompensated regulatory harms by protecting endowment effects
and other psychological connections between people and their property. It would constrain abuse by local governments, and it would be
consistent with many people’s deeply held intuitions. But there are
strong reasons to be skeptical of this prophylactic justification. Even
without a complete empirical cost-benefit accounting—an enterprise
far beyond the theoretical focus of this paper—it is clear that the costs
are likely to be too great and the benefits too small. This Part discusses these principle costs and benefits and then argues for more limited existing use protection.
277 For a sampling of this literature, see generally Mitchell N. Berman, Constitutional
Decision Rules, 90 VA. L. REV. 1, 10 (2004), which argues that decision rules are central to
judicially-created constitutional doctrine; Evan H. Caminker, Miranda and Some Puzzles
of “Prophylactic” Rules, 70 U. CIN. L. REV. 1, 9 (2001), which argues that where risk of
false negatives is high, a prophylactic rule might protect constitutional values; and David
A. Strauss, The Ubiquity of Prophylactic Rules, 55 U. CHI. L. REV. 190, 207–09 (1988),
which defends constitutional rules based on institutional propensities.
November 2009]
EXISTING USES
1283
A. The High Costs of Existing Use Protection
At a sufficient level of generality, the problem of existing uses
implicates an unexceptional application of the literature on legal transitions. Protecting owners from legal change—whether in the form of
compensation or grandfathering—creates predictable and potentially
perverse investment incentives. One of the most obvious is the opportunity for property owners to make investment decisions specifically
to take advantage of temporarily favorable regulatory treatment.278
In the face of a future regulatory change, property owners can lock in
a use by building or by vesting rights before the regulatory change
takes effect.279 This can precipitate a race to develop and can lead to
inefficient overinvestment in property.280 For just one example, in
GRA V, LLC v. Srinivasan,281 the New York Supreme Court
described the effect of neighborhood opposition to the development
of a new apartment building. According to the court, the result was a
race in which “the community sought to obtain a rezoning to prohibit
[the building], while [the] Owner endeavored to complete as much of
the construction as possible before any such rezoning.”282
There is a complicated but well-developed story to tell about
when, and under what conditions, protection from regulatory change
is likely to enhance efficiency.283 As Professor Shaviro has demonstrated, it depends on whether property owners should have an incentive to alter their conduct in anticipation of new rules.284 This, in turn,
can depend on whether the underlying rule change is good or not—
278 See Nash & Revesz, supra note 10, at 1725 (“[R]elief from a transition in legal
regimes is ordinarily inadvisable because it creates an incentive for societal actors not to
anticipate changes in the governing law.”); see also Lucas v. S.C. Coastal Council, 505 U.S.
1003, 1070 n.5 (Stevens, J., dissenting) (“In the face of uncertainty about changes in the
law, developers will overinvest, safe in the knowledge that if the law changes adversely,
they will be entitled to compensation.”); Serkin, supra note 222, at 677 & n.102 (citing
further sources and giving example).
279 Dana, supra note 40, at 677–81 (describing “race to develop”); see also Joseph
William Singer, The Ownership Society and Takings of Property: Castles, Investments, and
Just Obligations, 30 HARV. ENVTL. L. REV. 309, 325 (2006) (“[P]roperty owners . . . should
be forced to internalize [the risk of] . . . foreseeable new regulations designed to protect the
public from the harms attendant on the cumulative effects of individual acts of
ownership.”).
280 See supra Part III.A (discussing inefficiency in terms of antiretroactivity laws).
281 862 N.Y.S.2d 358 (App. Div. 2008).
282 Id. at 359–60.
283 In his careful taxonomy of retroactive effects, Professor Shaviro identifies instances
in which retroactive application of rules will not enhance efficiency. SHAVIRO, supra note
9, at 47–53 (looking specifically at retroactive application of taxes).
284 Id.; see also Logue, supra note 40, at 235–45 (identifying situations in which anticipation of legal change is desirable).
1284
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
that is, whether it represents legal progress, however defined.285 The
intuition is that individuals should adjust their behavior to anticipate
“good” rules but not “bad” ones.286 It can also depend on whether a
failure to anticipate some regulatory change will result in waste.287
There is little doubt, however, that protecting existing uses can induce
people to alter at least the timing of their investment decisions, if not
their substance, specifically in order to receive favorable regulatory
treatment.
Existing use protection can also lead to inefficient investment
decisions simply by allowing property owners to ignore the risk of regulatory change. Professor Kaplow examines this problem in his
leading article on legal transitions, arguing that compensation or other
protection for existing uses is likely to cause property owners to discount the risk of future regulation.288 If property owners know that
the government will compensate them for their property’s existing
uses, they may fail to account sufficiently for the risk of regulatory
change.289 According to Kaplow, this risk should be no different than
the risk of fire or flood; it too is a risk that property owners should at
least partially internalize in order to induce efficient levels of
investment.290
The effect of existing use protection on investment decisions is
obviously multifaceted and complex.291 At least for owners who are
285 SHAVIRO, supra note 9, at 48–51; see also Logue, supra note 40, at 236–39 (discussing
legal “disasters”).
286 See Logue, supra note 40, at 239–42 (“[I]f the transition is undesirable, the optimal
transition policy . . . would be a norm that actually discourages anticipation of the law
change and encourages reliance on old law . . . .”).
287 See id. at 236 (arguing that desirability of legal rule matters less when regulation
destroys property values entirely and irreversibly). For discussion of the problem of social
waste, see supra Part III.B.3.
288 Kaplow, supra note 9, at 529.
289 See id. at 529–31; see also Dana, supra note 40, at 679–80 (describing Kaplow’s
argument).
290 See Kaplow, supra note 9, at 528–30; see also SHAVIRO, supra note 9, at 49 (“Suppose I am considering building a house on a site where the government may at some point
exercise its power of eminent domain in order to extend a highway. When and if the
highway comes, should I be compensated for the value of my improvements? Here, my
incentives will be better if the answer is no.”); Levmore, supra note 40, at 1677–79 (“But if
the government pays for losses suffered as a result of regulation, then the factory owner
has no incentive to anticipate these regulations and to facilitate the move to better and
safer methods.”); Logue, supra note 40, at 237 (“To ensure that the individual has the ex
ante incentives (that is, to force her to internalize the cost of her decision to build on this
site), the consequentialist or economic framework would counsel against compensation
. . . .”).
291 It is, in fact, sufficiently complex that it demands considerably more treatment than it
can be given here. In particular, different regulatory contexts may afford different opportunities for owners to modify their behavior in anticipation of government action. This is a
topic that must be reserved for more comprehensive treatment in future work.
November 2009]
EXISTING USES
1285
risk averse, some kind of insurance is important for creating efficient
investment incentives.292 It may well be that individual homeowners
demand more protection against loss and are less susceptible to inefficient overinvestment than, say, owners of commercial or industrial
property, or developers, who may have greater ability to take strategic
advantage of existing use protection. Nevertheless, the possibility of
existing use protection creating perverse ex ante investment incentives in at least some regulatory contexts is of more than theoretical
concern. In the face of anticipated regulatory change, property
owners sometimes do, in fact, build aggressively to take advantage of
more lenient regulations.293 Even run-of-the-mill land use regulations
can stimulate more development, sooner, than would otherwise
occur.294
This is all relatively well-trodden ground in the takings and transitions literature: All else being equal, the categorical protection of
existing uses threatens to create perverse investment incentives for
property owners. But some costs are particularly pointed in the land
use context. For example, existing use protection can distort government decisionmaking. If existing uses are compensated, a government
sensitive to cost will avoid regulating property with existing uses.295
Knowing this, a property owner can, in effect, manipulate the government’s regulatory decisions by developing her property, making it
more costly for the government to regulate her property as opposed to
her neighbor’s undeveloped property.296 The protection of existing
uses may also make local governments less willing to experiment with
zoning. The less local governments retain the flexibility to get out of
unsuccessful zoning decisions, the less they will innovate where the
outcome is uncertain.297 This dynamic is also at least partly respon292 See, e.g., Doremus, supra note 200, at 16 (“Another efficiency concern is the worry
that an unstable regulatory climate will inhibit investment, particularly investment that
takes a long period of time to mature.”). In the tax context, Logue has argued persuasively
that failure to provide transition relief will result in taxpayers demanding a “default premium”—that is, more of a tax benefit in the future to induce such investments to account
for the possibility of retroactive change. See Logue, supra note 199, at 1139–41.
293 See, e.g., Dana, supra note 40, at 694–95 & n.107 (collecting examples of accelerated
development to beat regulatory clock).
294 For this reason, economist Robert Innes has argued that developed and undeveloped
property should receive equal treatment for takings purposes. See Robert Innes, Takings,
Compensation, and Equal Treatment for Owners of Developed and Undeveloped Property,
40 J.L. & ECON. 403, 406–07 (1997).
295 See Serkin, supra note 261, at 1666–73 (describing governments’ risk aversion).
296 See Innes, supra note 294, at 406 (“[T]he least valuable undeveloped land should be
taken first, which implies that, if takings are not compensated, landowners have an incentive to develop their land early in order to reduce their risk of government
appropriation.”).
297 I credit Professor Fischel with this interesting suggestion.
1286
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
sible for the recognition that zoning is not so much forward-looking
land use planning as it is mere description of existing conditions.298 If
planners have to zone around existing uses, then a significant part of
any zoning regime will incorporate the existing uses, whether or not
they make sense where they are.299 Categorical existing use protection can thereby hobble a local government’s zoning power by locking
in the status quo and preserving incompatible neighbors.
Finally, the political dynamic around existing uses suggests that
courts should be primarily concerned with the actions of small local
governments in which owners of existing uses suffer from systemic
political failures.300 This focus, however, threatens to exacerbate
some of the other costs of existing use protection. Small local governments are likely to be the most sensitive to litigation risks and to takings liability.301 In fact, they may be too sensitive and consequently
may be overdeterred from enacting what would have been beneficial
regulations.302 Also, the relatively low cost to developers and property owners of monitoring local government decisionmaking makes it
much easier to foresee adverse regulatory changes.303 Local government actions such as rezonings are therefore particularly likely to
create an inefficient race to invest in order to lock in existing
regulations.
At the very least, this catalog of costs calls into question the
appropriateness of existing use protection as a prophylactic rule. Legislatures in specific instances are likely to determine that the benefits
of protecting existing uses outweigh the costs, but categorically compelling the protection of existing uses eliminates both courts’ and legislatures’ ability to engage in any more fine-grained analysis of the
298 This concern is often voiced in the zoning and land use literature. See supra note 7
(citing sources).
299 Favoring existing uses over potential future uses may also privilege a particular kind
of property use. Criticisms of labor theories of property point out the harms of privileging
active uses of property over passive ones. See, e.g., Jedediah Purdy, Property and Empire:
The Law of Imperialism in Johnson v. M’Intosh, 75 GEO. WASH. L. REV. 329, 336 n.36
(2007) (describing Western accounts of property used to “justify colonial expropriations”).
The labor theory, in turn, has justified such pernicious acts as the expropriation of land by
early European settlers from Native Americans, whose use of land was not perceived as
being sufficiently active to warrant legal protection. See generally id. It may continue to
privilege buildings over open space, farmed land over conserved fields, and the like,
although perhaps the definition of “use” can be made sufficiently capacious to encompass
non-intensive uses such as conservation.
300 See supra Part III.D (discussing political economy of existing uses).
301 See Serkin, supra note 261, at 1666–67 (explaining why small local governments
might be more risk averse than larger governments).
302 See id. at 1666–67, 1672–74.
303 The costs of monitoring local governments are typically much lower than the costs of
monitoring larger governments. Id.
November 2009]
EXISTING USES
1287
relative costs and benefits of protecting versus regulating an existing
use in a specific context.
B. The Illusory Benefits of Existing Use Protection
Not only does existing use protection impose significant costs, but
many of its apparent benefits are largely illusory. Existing uses define
less of a bright line than one might generally suppose. There are two
principal problems. The first, quite simply, is found in the exceptions
to existing use protection. Because courts generally accept as inviolable the implicit protection of existing uses, they are forced to focus
on such theoretically unsatisfying questions as whether pouring a
footing for a foundation is sufficient to vest development rights304 or
whether environmental protection counts as preventing a harm or
conferring a benefit.305 If the supposed benefit of categorical existing
use protection is the ease with which it can be applied, its myriad and
murky exceptions create a doctrine that is actually difficult to apply
and theoretically unsatisfying. Focusing, for example, on the development necessary to create a vested right obscures a more straightforward inquiry into the costs and benefits of applying a specific
regulation to a particular property, existing use or not.
The second problem with existing uses as a category is deciding
what it actually includes. If a local government regulates away an
existing adult bookstore but still allows commercial use of the property, has it eliminated a bookstore or has it created some new regulatory restrictions on an existing and ongoing commercial enterprise?
How the use is characterized will determine whether or not a regulation even implicates the protection of an existing use. The more narrowly an existing use is defined, the more any regulation will look like
it eliminates that use. To take an extreme example, a regulation
requiring landlords to install smoke detectors could be construed as
eliminating an existing use, if the use is defined as an apartment
building without smoke detectors. Similarly, if a local government
requires homeowners with wells to hook into the municipal water
supply, it could be seen as eliminating the existing well or merely as
regulating the continuing use of the house.306 Existing uses may not
304 See supra notes 76–82 and accompanying text (discussing Prince George’s County v.
Sunrise Development Ltd. Partnership, 623 A.2d 1296 (Md. 1993)).
305 See supra note 92 and accompanying text (noting that distinction between preventing
harm and conferring benefit is unsettled).
306 See City of Gaylord v. Maple Manor Invs., No. 266954, 2006 WL 2270494, at *5
(Mich. Ct. App. Aug. 8, 2006) (finding that home rule cities have authority, under police
power, to require homeowners to connect to city’s sewer system and cease use of wells).
1288
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
define an easily administrable category for property protection after
all.
C. Designing Appropriate Protection for Existing Uses
The normative justifications for current existing use protection
are surprisingly unconvincing. The costs are substantial. This is not to
suggest, however, that existing uses are irrelevant and should simply
be ignored. Rather, it is to argue that blanket, categorical rules
exempting existing uses from regulatory change protect too much.
Courts and government actors should instead apply the same kind of
inquiry to both existing and prospective future uses. The same inquiry
may well generate a different answer if a use already exists, leading to
different results, but not because there is anything sacrosanct about
the existing use itself.
This is an important, but still general conclusion. What does it
mean on the ground? Perhaps most important is what it does not
mean. Following this approach, governments still should not make a
common practice of trampling over existing uses. In many—if not
most—instances, governments should protect existing uses as part of
the usual cost-benefit analysis that they undertake.307 If in a particular situation eliminating existing uses will impose more costs than
benefits, then the government should not eliminate them. A typical
analysis should take into account the costs of retrofitting existing uses,
distributive consequences, and the ultimate effect of existing use protection.308 Ideally, the government will include in its assessment some
accounting for property owners’ connections to their property. This,
however, is not fundamentally different from any government decisionmaking that involves weighing competing costs and benefits from
various sources, and it is ultimately an empirical question whether a
particular existing use should be protected in a particular case.309 In
other contexts involving economic regulations, courts rarely interfere
307 For a thoroughgoing discussion of sophisticated cost-benefit analysis in government
decisionmaking, see generally RICHARD L. REVESZ & MICHAEL A. LIVERMORE,
RETAKING RATIONALITY: HOW COST-BENEFIT ANALYSIS CAN BETTER PROTECT THE
ENVIRONMENT AND OUR HEALTH (2008).
308 For a sophisticated analysis of the systemic effect of grandfathering provisions in
environmental regulations, see generally Nash & Revesz, supra note 10. For a formal
model developed by Shavell to evaluate the effect of grandfathering provisions generally,
see Shavell, supra note 17, at 57–67.
309 See Nash & Revesz, supra note 10, at 1712 (“In general, the question whether
grandfathering combined with more stringent regulation of new sources will lead to more
pollution is an empirical one.”). There are, of course, competing and less rational accounts
of how governments actually make decisions. See supra note 261 and accompanying text
(discussing public choice theory).
November 2009]
EXISTING USES
1289
to second-guess those cost-benefit determinations.310 There is little, if
anything, conceptually different about existing uses that justifies special judicial protection.
This Article’s conclusion also must not be taken to suggest that
courts should abdicate any role in reviewing government regulations
of existing uses. Indeed, regulations of existing uses should be subject
to the same kinds of takings and due process analysis that applies to
all government actions. If a regulation goes too far—if it results in too
great a diminution in value, or if it is arbitrary or irrational—then
courts should strike it down.311 As discussed above, regulations of
existing uses are more likely to trigger takings liability under existing
takings doctrine than regulations of prospective future uses because
the economic impact is likely to be greater.312 But courts should not
extend special protection to existing uses, let alone the kind of incredibly strong protection that current land use doctrines provide.
Consider a few examples. Imagine a local government seeking to
apply new minimum lot size requirements to an existing house on a
one-acre lot. If the new zoning ordinance only permits one house
every two acres, rendering the house a prior nonconforming use, the
local government should not force the property owner to tear down
her house to come into conformity. The costs of doing so would be
substantial and would include not only the demolition costs but also
the owner’s subjective value in her property. If a local government
nevertheless tried to impose such a regulation on existing property, a
court should invalidate it under the Takings Clause.
Contrast this example with a new ordinance regulating the placement of billboards on roadsides. Its application to existing billboards
may well come out quite differently. Of course, if the ordinance was
enacted to prevent traffic accidents, it could fall under the harmprevention exception and be permissible even under current law.313
But if the purpose was instead scenic preservation, or something less
clearly implicating health, safety, and welfare, should the government
nevertheless be able to act? Here, if the costs of moving the billboards are sufficiently low and the benefits of moving them sufficiently high, it may well be appropriate to apply the ordinance to
existing billboards. After all, grandfathering existing billboards runs
the risk of creating a race to build billboards in precisely those places
310
See supra note 181 (citing sources discussing substantive due process).
These are summaries of the leading takings and due process tests, discussed in Part
II, supra.
312 See supra note 168 (citing sources discussing economic substantive due process).
313 See supra Part I.D (discussing nuisance exception).
311
1290
NEW YORK UNIVERSITY LAW REVIEW
[Vol. 84:1222
where they will become impermissible.314 This will undermine the
efficacy of the ordinance and will have the perverse effect of conferring extra value on those nonconforming billboards as a direct result
of their nonconforming status. It should be much harder for a court
reviewing the application of this ordinance to existing uses to find a
constitutional violation. Especially if the property can still be used for
billboards—so long as they are placed in a different location—the
effect on the value of the property is likely to be negligible. Similarly,
so long as the costs of moving the billboards are sufficiently low, there
would not appear to be anything irrational or arbitrary about applying
the ordinance to existing billboards. To the contrary, doing so would
increase the ordinance’s efficacy and thus be perfectly rational.
It is, of course, the cases in the middle that will prove most
vexing. Eliminating sheds or out-buildings, forbidding the ongoing
use of well-water or septic systems and requiring hook-ups to municipal infrastructure instead, or regulating away a cellular tower all present more complex cost-benefit tradeoffs. But that, ultimately, is the
point. Once existing uses are dismantled as a specially protected category, it is possible to address whether a particular existing use should
be protected with more precision and sophistication about the costs
and benefits actually at stake.
At the end of the day, it is this final point that is probably the
most important. With the background rule of existing use protection,
the difficult doctrinal work in the area is channeled into the rule’s
numerous but often peculiar exceptions. This, in turn, puts enormous
pressure on such amorphous tests as vested rights, the harm exception
for takings liability, and the minimum duration of amortization provisions. These tests do nothing to advance the clarity of land use doctrine and actually obstruct a more direct inquiry into the character and
magnitude of a regulation’s effects on property rights.
CONCLUSION
By eliminating the special protection of existing uses, legislatures
and courts would be able to engage in a more fine-grained and factspecific analysis when deciding whether the government can eliminate
an existing use. Instead of creating a sweeping prohibition, courts
would give more latitude to regulatory strategies that interfere with
existing uses in particular cases. At the same time, this Article articulates for governments the kinds of costs and concerns that existing
uses present. It opens to governments the possibility of regulating
existing uses more often than current law allows while highlighting the
314
See supra Part IV.A (discussing incentives created by grandfathering).
November 2009]
EXISTING USES
1291
reasons for caution. Ultimately, though, that caution is best exercised
by government decisionmakers instead of by courts prohibiting a
whole category of regulations based on vague constitutional principles
and thin normative justifications.
Many areas of law—from the Takings Clause, to various state and
federal statutes, to a variety of land use doctrines—provide special
protection for existing uses. The current assumptions that existing
uses are entitled to strong if not absolute protection should be eliminated. Existing uses should, instead, be subject to the same constitutional protection as prospective future uses. This will result in greater
freedom for government actors to choose for themselves when,
whether, and how to treat existing uses when enacting a regulatory
change.